2022/01/22 更新

写真a

オガタ テツヤ
尾形 哲也
所属
理工学術院 基幹理工学部
職名
教授
メールアドレス
メールアドレス
プロフィール
1993年早稲田大学理工学部機械工学科卒業.1997年日本学術振興会特別研究員(DC2),1999年早稲田大学理工学部助手,2001年理化学研究所脳科学総合研究センター研究員,2003年京都大学大学院情報学研究科講師,2005年同准教授を経て,2012年より早稲田大学理工学術院教授.博士(工学).2009年-2015年JSTさきがけ領域研究員,また2017年より産業総合技術研究所人工知能研究センター特定フェロー.2013年から2014年日本ロボット学会理事,2016年から2018年人工知能学会理事などを歴任.2017年より日本ディープラーニング協会理事.ディープラーニングに代表される神経回路モデルとロボットシステムを用いた,認知ロボティクス研究,特に予測学習,模倣学習,マルチモーダル統合,言語学習,コミュニケーションなどの研究に従事.

兼担

  • 附属機関・学校   グローバルエデュケーションセンター

  • 理工学術院   大学院基幹理工学研究科

学内研究所等

  • 2021年
    -
    2022年

    データ科学センター   兼任センター員

  • 2020年
    -
    2022年

    理工学術院総合研究所   兼任研究員

  • 2020年
    -
    2022年

    リサーチイノベ オープンイノベーション推進セクション   兼任センター員

学歴

  • 1995年04月
    -
    1998年03月

    早稲田大学   大学院理工学研究科   機械工学専攻  

    博士後期課程

  • 1993年04月
    -
    1995年03月

    早稲田大学   大学院理工学研究科   機械工学専攻  

    修士課程

  • 1989年04月
    -
    1993年03月

    早稲田大学   理工学部   機械工学科  

学位

  • 2000年03月   早稲田大学   博士(工学)

経歴

  • 2017年10月
    -
    継続中

    産業技術総合研究所   人工知能研究センター   特定フェロー

  • 2012年04月
    -
    継続中

    早稲田大学   理工学術院   教授

  • 2009年10月
    -
    2015年03月

    科学技術振興機構   さきがけ領域研究員

  • 2005年06月
    -
    2012年03月

    京都大学   情報学研究科   准教授

  • 2003年10月
    -
    2005年05月

    京都大学   情報学研究科   講師

  • 2001年04月
    -
    2003年09月

    理化学研究所   脳科学総合研究センター   研究員

  • 1999年04月
    -
    2001年03月

    早稲田大学   理工学部   助手

  • 1997年04月
    -
    1999年03月

    日本学術振興会   特別研究員

▼全件表示

所属学協会

  •  
     
     

    IEEE

  •  
     
     

    人工知能学会

  •  
     
     

    計測自動制御学会

  •  
     
     

    バイオメカニズム学会

  •  
     
     

    日本機械学会

  •  
     
     

    日本ロボット学会

  •  
     
     

    情報処理学会

▼全件表示

 

研究分野

  • 知能ロボティクス

研究キーワード

  • 深層予測学習

  • 認知ロボティクス

論文

  • Utilization of Image/Force/Tactile Sensor Data for Object-Shape-Oriented Manipulation: Wiping Objects With Turning Back Motions and Occlusion.

    Namiko Saito, Takumi Shimizu, Tetsuya Ogata, Shigeki Sugano

    IEEE Robotics and Automation Letters   7 ( 2 ) 968 - 975  2022年

    DOI

  • Leveraging motor babbling for efficient robot learning

    Kei Kase, Noboru Matsumoto, Tetsuya Ogata

    Journal of Robotics and Mechatronics   33 ( 5 ) 1063 - 1074  2021年10月

     概要を見る

    Deep robotic learning by learning from demonstration allows robots to mimic a given demonstration and generalize their performance to unknown task setups. However, this generalization ability is heavily affected by the number of demonstrations, which can be costly to manually generate. Without sufficient demonstrations, robots tend to overfit to the available demonstrations and lose the robustness offered by deep learning. Applying the concept of motor babbling – a process similar to that by which human infants move their bodies randomly to obtain proprioception – is also effective for allowing robots to enhance their generalization ability. Furthermore, the generation of babbling data is simpler than task-oriented demonstrations. Previous researches use motor babbling in the concept of pre-training and fine-tuning but have the problem of the babbling data being overwritten by the task data. In this work, we propose an RNNbased robot-control framework capable of leveraging targetless babbling data to aid the robot in acquiring proprioception and increasing the generalization ability of the learned task data by learning both babbling and task data simultaneously. Through simultaneous learning, our framework can use the dynamics obtained from babbling data to learn the target task efficiently. In the experiment, we prepare demonstrations of a block-picking task and aimless-babbling data. With our framework, the robot can learn tasks faster and show greater generalization ability when blocks are at unknown positions or move during execution.

    DOI

  • Tool-Use Model to Reproduce the Goal Situations Considering Relationship Among Tools, Objects, Actions and Effects Using Multimodal Deep Neural Networks

    Namiko Saito, Tetsuya Ogata, Hiroki Mori, Shingo Murata, Shigeki Sugano

    Frontiers in Robotics and AI   8  2021年09月

     概要を見る

    We propose a tool-use model that enables a robot to act toward a provided goal. It is important to consider features of the four factors; tools, objects actions, and effects at the same time because they are related to each other and one factor can influence the others. The tool-use model is constructed with deep neural networks (DNNs) using multimodal sensorimotor data; image, force, and joint angle information. To allow the robot to learn tool-use, we collect training data by controlling the robot to perform various object operations using several tools with multiple actions that leads different effects. Then the tool-use model is thereby trained and learns sensorimotor coordination and acquires relationships among tools, objects, actions and effects in its latent space. We can give the robot a task goal by providing an image showing the target placement and orientation of the object. Using the goal image with the tool-use model, the robot detects the features of tools and objects, and determines how to act to reproduce the target effects automatically. Then the robot generates actions adjusting to the real time situations even though the tools and objects are unknown and more complicated than trained ones.

    DOI

  • Paradoxical sensory reactivity induced by functional disconnection in a robot model of neurodevelopmental disorder

    Hayato Idei, Shingo Murata, Yuichi Yamashita, Tetsuya Ogata

    Neural Networks   138   150 - 163  2021年06月  [査読有り]

     概要を見る

    Neurodevelopmental disorders are characterized by heterogeneous and non-specific nature of their clinical symptoms. In particular, hyper- and hypo-reactivity to sensory stimuli are diagnostic features of autism spectrum disorder and are reported across many neurodevelopmental disorders. However, computational mechanisms underlying the unusual paradoxical behaviors remain unclear. In this study, using a robot controlled by a hierarchical recurrent neural network model with predictive processing and learning mechanism, we simulated how functional disconnection altered the learning process and subsequent behavioral reactivity to environmental change. The results show that, through the learning process, long-range functional disconnection between distinct network levels could simultaneously lower the precision of sensory information and higher-level prediction. The alteration caused a robot to exhibit sensory-dominated and sensory-ignoring behaviors ascribed to sensory hyper- and hypo-reactivity, respectively. As long-range functional disconnection became more severe, a frequency shift from hyporeactivity to hyperreactivity was observed, paralleling an early sign of autism spectrum disorder. Furthermore, local functional disconnection at the level of sensory processing similarly induced hyporeactivity due to low sensory precision. These findings suggest a computational explanation for paradoxical sensory behaviors in neurodevelopmental disorders, such as coexisting hyper- and hypo-reactivity to sensory stimulus. A neurorobotics approach may be useful for bridging various levels of understanding in neurodevelopmental disorders and providing insights into mechanisms underlying complex clinical symptoms.

    DOI PubMed

  • Development of a Basic Educational Kit for Robotic System with Deep Neural Networks

    Momomi Kanamura, Kanata Suzuki, Yuki Suga, Tetsuya Ogata

    Sensors   21 ( 11 ) 3804 - 3804  2021年05月  [査読有り]

    担当区分:責任著者

     概要を見る

    In many robotics studies, deep neural networks (DNNs) are being actively studied due to their good performance. However, existing robotic techniques and DNNs have not been systematically integrated, and packages for beginners are yet to be developed. In this study, we proposed a basic educational kit for robotic system development with DNNs. Our goal was to educate beginners in both robotics and machine learning, especially the use of DNNs. Initially, we required the kit to (1) be easy to understand, (2) employ experience-based learning, and (3) be applicable in many areas. To clarify the learning objectives and important parts of the basic educational kit, we analyzed the research and development (R&D) of DNNs and divided the process into three steps of data collection (DC), machine learning (ML), and task execution (TE). These steps were configured under a hierarchical system flow with the ability to be executed individually at the development stage. To evaluate the practicality of the proposed system flow, we implemented it for a physical robotic grasping system using robotics middleware. We also demonstrated that the proposed system can be effectively applied to other hardware, sensor inputs, and robot tasks.

    DOI PubMed

  • From Anime To Reality: Embodying An Anime Character As A Humanoid Robot

    Mohammed Al Al Sada, Pin-Chu Yang, Chang Chieh Chiu, Tito Pradhono Tomo, Mhd Yamen Saraiji, Tetsuya Ogata, Tatsuo Nakajima

    Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems    2021年05月  [査読有り]

    DOI

  • Embodying Pre-Trained Word Embeddings Through Robot Actions

    Minori Toyoda, Kanata Suzuki, Hiroki Mori, Yoshihiko Hayashi, Tetsuya Ogata

    IEEE Robotics and Automation Letters   6 ( 2 ) 4225 - 4232  2021年04月  [査読有り]

    担当区分:責任著者

    DOI

  • Compensation for Undefined Behaviors during Robot Task Execution by Switching Controllers Depending on Embedded Dynamics in RNN

    Kanata Suzuki, Hiroki Mori, Tetsuya Ogata

    IEEE Robotics and Automation Letters   6 ( 2 ) 3475 - 3482  2021年04月  [査読有り]

    担当区分:最終著者, 責任著者

     概要を見る

    Robotic applications require both correct task performance and compensation for undefined behaviors. Although deep learning is a promising approach to perform complex tasks, the response to undefined behaviors that are not reflected in the training dataset remains challenging. In a human-robot collaborative task, the robot may adopt an unexpected posture due to collisions and other unexpected events. Therefore, robots should be able to recover from disturbances for completing the execution of the intended task. We propose a compensation method for undefined behaviors by switching between two controllers. Specifically, the proposed method switches between learning-based and model-based controllers depending on the internal representation of a recurrent neural network that learns task dynamics. We applied the proposed method to a pick-And-place task and evaluated the compensation for undefined behaviors. Experimental results from simulations and on a real robot demonstrate the effectiveness and high performance of the proposed method.

    DOI

  • How to Select and Use Tools? : Active Perception of Target Objects Using Multimodal Deep Learning

    Namiko Saito, Tetsuya Ogata, Satoshi Funabashi, Hiroki Mori, Shigeki Sugano

    IEEE Robotics and Automation Letters   6 ( 2 ) 2517 - 2524  2021年04月  [査読有り]

     概要を見る

    Selection of appropriate tools and use of them when performing daily tasks is a critical function for introducing robots for domestic applications. In previous studies, however, adaptability to target objects was limited, making it difficult to accordingly change tools and adjust actions. To manipulate various objects with tools, robots must both understand tool functions and recognize object characteristics to discern a tool-object-action relation. We focus on active perception using multimodal sensorimotor data while a robot interacts with objects, and allow the robot to recognize their extrinsic and intrinsic characteristics. We construct a deep neural networks (DNN) model that learns to recognize object characteristics, acquires tool-object-action relations, and generates motions for tool selection and handling. As an example tool-use situation, the robot performs an ingredients transfer task, using a turner or ladle to transfer an ingredient from a pot to a bowl. The results confirm that the robot recognizes object characteristics and servings even when the target ingredients are unknown. We also examine the contributions of images, force, and tactile data and show that learning a variety of multimodal information results in rich perception for tool use.

    DOI

  • Viewpoint Planning Based on Uncertainty Maps Created from the Generative Query Network

    Kelvin Lukman, Hiroki Mori, Tetsuya Ogata

    Advances in Intelligent Systems and Computing     37 - 48  2021年

    DOI

  • In-air Knotting of Rope using Dual-Arm Robot based on Deep Learning.

    Kanata Suzuki, Momomi Kanamura, Yuki Suga, Hiroki Mori, Tetsuya Ogata

    CoRR   abs/2103.09402  2021年

  • Spatial Attention Point Network for Deep-learning-based Robust Autonomous Robot Motion Generation.

    Hideyuki Ichiwara, Hiroshi Ito, Kenjiro Yamamoto, Hiroki Mori, Tetsuya Ogata

    CoRR   abs/2103.01598  2021年

  • Stable deep reinforcement learning method by predicting uncertainty in rewards as a subtask.

    Kanata Suzuki, Tetsuya Ogata

    CoRR   abs/2101.06906  2021年

  • Tactile-based curiosity maximizes tactile-rich object-oriented actions even without any extrinsic rewards

    Hiroki Mori, Masayuki Masuda, Tetsuya Ogata

    ICDL-EpiRob 2020 - 10th IEEE International Conference on Development and Learning and Epigenetic Robotics    2020年10月  [査読有り]

     概要を見る

    This study proposed a hypothesis regarding the emergence of object-oriented action via tactile-based curiosity. The hypothesis is such that a curious exploration driven by tactile sensation leads tactile-rich object-oriented actions, while there are no explicit rewards or other designated intentional purposes. Experiments were with the curiosity model named the disagreement model from the reinforcement learning research field and with a simple physics robotic simulation with visual and tactile sensory information. The experimental results indicated that the tactile sensation induces object-oriented actions such as hitting and pecking by the body parts that have tactile sensors. We deduced that the hypothesis could be extended to discussions regarding the acquisition of dexterous skillful object manipulation in human development.

    DOI

  • Wiping 3D-objects using deep learning model based on image/force/joint information

    Namiko Saito, Danyang Wang, Tetsuya Ogata, Hiroki Mori, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems     10152 - 10157  2020年10月  [査読有り]

     概要を見る

    We propose a deep learning model for a robot to wipe 3D-objects. Wiping of 3D-objects requires recognizing the shapes of objects and planning the motor angle adjustments for tracing the objects. Unlike previous research, our learning model does not require pre-designed computational models of target objects. The robot is able to wipe the objects to be placed by using image, force, and arm joint information. We evaluate the generalization ability of the model by confirming that the robot handles untrained cube and bowl shaped-objects. We also find that it is necessary to use both image and force information to recognize the shape of and wipe 3D objects consistently by comparing changes in the input sensor data to the model. To our knowledge, this is the first work enabling a robot to use learning sensorimotor information alone to trace various unknown 3D-shape.

    DOI

  • Variable in-hand manipulations for tactile-driven robot hand via CNN-LSTM

    Satoshi Funabashi, Shun Ogasa, Tomoki Isobe, Tetsuya Ogata, Alexander Schmitz, Tito Pradhono Tomo, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems     9472 - 9479  2020年10月  [査読有り]

     概要を見る

    Performing various in-hand manipulation tasks, without learning each individual task, would enable robots to act more versatile, while reducing the effort for training. However, in general it is difficult to achieve stable in-hand manipulation, because the contact state between the fingertips becomes difficult to model, especially for a robot hand with anthropomorphically shaped fingertips. Rich tactile feedback can aid the robust task execution, but on the other hand it is challenging to process high-dimensional tactile information. In the current paper we use two fingers of the Allegro hand, and each fingertip is anthropomorphically shaped and equipped not only with 6-axis force-torque (F/T) sensors, but also with uSkin tactile sensors, which provide 24 tri-axial measurements per fingertip. A convolutional neural network is used to process the high dimensional uSkin information, and a long short-term memory (LSTM) handles the time-series information. The network is trained to generate two different motions ("twist"and "push"). The desired motion is provided as a task-parameter to the network, with twist defined as -1 and push as +1. When values between -1 and +1 are used as the task parameter, the network is able to generate untrained motions in-between the two trained motions. Thereby, we can achieve multiple untrained manipulations, and can achieve robustness with high-dimensional tactile feedback.

    DOI

  • Stable in-grasp manipulation with a low-cost robot hand by using 3-axis tactile sensors with a CNN

    Satoshi Funabashi, Tomoki Isobe, Shun Ogasa, Tetsuya Ogata, Alexander Schmitz, Tito Pradhono Tomo, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems     9166 - 9173  2020年10月  [査読有り]

     概要を見る

    The use of tactile information is one of the most important factors for achieving stable in-grasp manipulation. Especially with low-cost robotic hands that provide low-precision control, robust in-grasp manipulation is challenging. Abundant tactile information could provide the required feed-back to achieve reliable in-grasp manipulation also in such cases. In this research, soft distributed 3-axis skin sensors ("uSkin") and 6-axis F/T (force/torque) sensors were mounted on each fingertip of an Allegro Hand to provide rich tactile information. These sensors yielded 78 measurements for each fingertip (72 measurements from the uSkin and 6 measurements from the 6-axis F/T sensor). However, such high-dimensional tactile information can be difficult to process because of the complex contact states between the grasped object and the fingertips. Therefore, a convolutional neural network (CNN) was employed to process the tactile information. In this paper, we explored the importance of the different sensors for achieving in-grasp manipulation. Successful in-grasp manipulation with untrained daily objects was achieved when both 3-axis uSkin and 6-axis F/T information was provided and when the information was processed using a CNN.

    DOI

  • Homogeneous Intrinsic Neuronal Excitability Induces Overfitting to Sensory Noise: A Robot Model of Neurodevelopmental Disorder

    Hayato Idei, Shingo Murata, Yuichi Yamashita, Tetsuya Ogata

    Frontiers in Psychiatry   11  2020年08月  [査読有り]

     概要を見る

    Neurodevelopmental disorders, including autism spectrum disorder, have been intensively investigated at the neural, cognitive, and behavioral levels, but the accumulated knowledge remains fragmented. In particular, developmental learning aspects of symptoms and interactions with the physical environment remain largely unexplored in computational modeling studies, although a leading computational theory has posited associations between psychiatric symptoms and an unusual estimation of information uncertainty (precision), which is an essential aspect of the real world and is estimated through learning processes. Here, we propose a mechanistic explanation that unifies the disparate observations via a hierarchical predictive coding and developmental learning framework, which is demonstrated in experiments using a neural network-controlled robot. The results show that, through the developmental learning process, homogeneous intrinsic neuronal excitability at the neural level induced via self-organization changes at the information processing level, such as hyper sensory precision and overfitting to sensory noise. These changes led to multifaceted alterations at the behavioral level, such as inflexibility, reduced generalization, and motor clumsiness. In addition, these behavioral alterations were accompanied by fluctuating neural activity and excessive development of synaptic connections. These findings might bridge various levels of understandings in autism spectrum and other neurodevelopmental disorders and provide insights into the disease processes underlying observed behaviors and brain activities in individual patients. This study shows the potential of neurorobotics frameworks for modeling how psychiatric disorders arise from dynamic interactions among the brain, body, and uncertain environments.

    DOI

  • HATSUKI : An anime character like robot figure platform with anime-style expressions and imitation learning based action generation

    Pin Chu Yang, Mohammed Al-Sada, Chang Chieh Chiu, Kevin Kuo, Tito Pradhono Tomo, Kanata Suzuki, Nelson Yalta, Kuo Hao Shu, Tetsuya Ogata

    29th IEEE International Conference on Robot and Human Interactive Communication, RO-MAN 2020     384 - 391  2020年08月  [査読有り]

     概要を見る

    Japanese character figurines are popular and have a pivot position in Otaku culture. Although numerous robots have been developed, few have focused on otaku-culture or on embodying anime character figurines. Therefore, we take the first steps to bridge this gap by developing Hatsuki, which is a humanoid robot platform with anime based design. Hatsuki's novelty lies in its aesthetic design, 2D facial expressions, and anime-style behaviors that allows Hatsuki to deliver rich interaction experiences resembling anime-characters. We explain our design implementation process of Hatsuki, followed by our evaluations. In order to explore user impressions and opinions towards Hatsuki, we conducted a questionnaire in the world's largest anime-figurine event. The results indicate that participants were generally very satisfied with Hatsuki's design, and proposed various use case scenarios and deployment contexts for Hatsuki. The second evaluation focused on imitation learning, as such a method can provide better interaction ability in the real world and generate rich, context-adaptive behaviors in different situations. We made Hatsuki learn 11 actions, combining voice, facial expressions and motions, through the neural network based policy model with our proposed interface. Results show our approach was successfully able to generate the actions through self-organized contexts, which shows the potential for generalizing our approach in further actions under different contexts. Lastly, we present our future research direction for Hatsuki and provide our conclusion.

    DOI

  • 未知語に対応可能な言語と動作の統合表現獲得モデル

    豊田みのり, 森裕紀, 鈴木彼方, 林良彦, 尾形哲也

    第34回人工知能学会全国大会     2D4-OS-18a-04  2020年06月

  • Transferable Task Execution from Pixels through Deep Planning Domain Learning

    Kei Kase, Chris Paxton, Hammad Mazhar, Tetsuya Ogata, Dieter Fox

    Proceedings - IEEE International Conference on Robotics and Automation     10459 - 10465  2020年05月  [査読有り]

     概要を見る

    While robots can learn models to solve many manipulation tasks from raw visual input, they cannot usually use these models to solve new problems. On the other hand, symbolic planning methods such as STRIPS have long been able to solve new problems given only a domain definition and a symbolic goal, but these approaches often struggle on the real world robotic tasks due to the challenges of grounding these symbols from sensor data in a partially-observable world. We propose Deep Planning Domain Learning (DPDL), an approach that combines the strengths of both methods to learn a hierarchical model. DPDL learns a high-level model which predicts values for a large set of logical predicates consisting of the current symbolic world state, and separately learns a low-level policy which translates symbolic operators into executable actions on the robot. This allows us to perform complex, multistep tasks even when the robot has not been explicitly trained on them. We show our method on manipulation tasks in a photorealistic kitchen scenario.

    DOI

  • 料理ロボットのための道具の選択・使用深層学習モデル – 道具と食材の配置に応じた料理のよそい動作の実現

    斎藤菜美子, 呉雨恒, 尾形哲也, 森裕紀, 王丹阳, 陽品駒, 菅野重樹

    情報処理学会第82回全国大会     5U-08  2020年03月

  • 位置情報を明示的に扱う空間的注意機構モデルによる物体位置・角度推定の汎化

    昼間彪吾, 森裕紀, 尾形哲也

    情報処理学会第82回全国大会    2020年03月

  • Evaluation of Generalization Performance of Visuo-Motor Learning by Analyzing Internal State Structured from Robot Motion

    Hiroshi Ito, Kenjiro Yamamoto, Hiroki Mori, Tetsuya Ogata

    New Generation Computing   38 ( 1 ) 7 - 22  2020年03月  [査読有り]

     概要を見る

    To provide robots with the flexibility they need to cope with various environments, motion generation techniques using deep learning have been proposed. Generalization in deep learning is expected to enable flexible processing in unknown situations and flexible motion generation. Motion generation models have been proposed to realize specific robot tasks, and their operation successes in unknown situations have been reported. However, their generalization performances have not been analyzed or verified in detail. In this paper, we analyze the internal state of a deep neural network using principal component analysis and verify the generalization of motion against environmental change, specifically a repositioned door knob. The results revealed that the motion primitives were structured in accordance with the position of the door knob. In addition, motion with high generalization performance was obtained by adaptive transition of motion primitives in accordance with changes in the door knob position. The robot was able to successfully perform a door-open-close task at various door knob positions.

    DOI

  • 不確実性の推定と自閉スペクトラム症ー神経ロボティクス実験による症状シミュレーションー

    出井勇人, 村田真悟, 尾形哲也, 山下祐一

    精神医学   62 ( 2 ) 219 - 229  2020年02月  [査読有り]

  • HATSUKI : An anime character like robot figure platform with anime-style expressions and imitation learning based action generation.

    Pin-Chu Yang, Mohammed Al-Sada, Chang-Chieh Chiu, Kevin Kuo, Tito Pradhono Tomo, Kanata Suzuki, Nelson Yalta, Kuo-Hao Shu, Tetsuya Ogata

    CoRR   abs/2003.14121  2020年

  • Transferable Task Execution from Pixels through Deep Planning Domain Learning.

    Kei Kase, Chris Paxton, Hammad Mazhar, Tetsuya Ogata, Dieter Fox

    CoRR   abs/2003.03726  2020年

  • Development of a Basic Educational Kit for Robot Development Using Deep Neural Networks

    Momomi Kanamura, Yuki Suga, Tetsuya Ogata

    Proceedings of the 2020 IEEE/SICE International Symposium on System Integration, SII 2020     1360 - 1363  2020年01月  [査読有り]

     概要を見る

    In this paper, we propose a basic educational kit for robotic system development using deep neural networks (DNNs). To develop systems robust to changes in dynamic environments, much research is focusing on learning-based recognition and robotic manipulation systems. However, existing robotic techniques and DNNs are not systematically integrated and packages for beginners have not yet been developed. Therefore, we developed a robotic system using DNNs and a system manual to serve as a basic educational kit that can be easily used by anyone. Our goal was to educate beginners in both robotics and machine learning, especially when DNNs are used. Initially, we set the following requirements for the kit: (1) easy to understand; (2) experience-based; and (3) applicable in many areas. We analyzed the research and development of DNNs and divided the process into Date Collecting (DC), Machine Learning (ML), and Task Execution (TE) steps. Finally, we applied our hierarchical system architecture to a physical robotic grasping system. We implemented the DC and TE steps in the physical robotic system by using robotics middleware and in the ML step, we prepared a sample script.

    DOI

  • Stable Deep Reinforcement Learning Method by Predicting Uncertainty in Rewards as a Subtask

    Kanata Suzuki, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   12533 LNCS   651 - 662  2020年  [査読有り]

     概要を見る

    In recent years, a variety of tasks have been accomplished by deep reinforcement learning (DRL). However, when applying DRL to tasks in a real-world environment, designing an appropriate reward is difficult. Rewards obtained via actual hardware sensors may include noise, misinterpretation, or failed observations. The learning instability caused by these unstable signals is a problem that remains to be solved in DRL. In this work, we propose an approach that extends existing DRL models by adding a subtask to directly estimate the variance contained in the reward signal. The model then takes the feature map learned by the subtask in a critic network and sends it to the actor network. This enables stable learning that is robust to the effects of potential noise. The results of experiments in the Atari game domain with unstable reward signals show that our method stabilizes training convergence. We also discuss the extensibility of the model by visualizing feature maps. This approach has the potential to make DRL more practical for use in noisy, real-world scenarios.

    DOI

  • Visualization of Focal Cues for Visuomotor Coordination by Gradient-based Methods: A Recurrent Neural Network Shifts the Attention Depending on Task Requirements

    Hiroshi Ito, Kenjiro Yamamoto, Hiroki Mori, Shuki Goto, Tetsuya Ogata

    Proceedings of the 2020 IEEE/SICE International Symposium on System Integration, SII 2020     188 - 194  2020年01月  [査読有り]

     概要を見る

    For an autonomous robot to flexibly move in response to various tasks or environmental changes, an attention mechanism is required that is based on the robot's behavioral experience. In this paper, we visualize how attention is acquired inside a neural network learned using supervised learning and describe how to acquire a suitable representation for performing a task. Our experimental evaluation shows that the attention was automatically acquired for objects that are needed to perform tasks by learning the time-series of both vision and motor information rather than only vision information. By multimodal learning, the attention is robust against unlearned conditions which background changes or obstacles.

    DOI

  • Real-time liquid pouring motion generation: End-to-end sensorimotor coordination for unknown liquid dynamics trained with deep neural networks

    Namiko Saito, Nguyen Ba Dai, Tetsuya Ogata, Hiroki Mori, Shigeki Sugano

    IEEE International Conference on Robotics and Biomimetics, ROBIO 2019     1077 - 1082  2019年12月  [査読有り]

     概要を見る

    We propose a sensorimotor dynamical system model for pouring unknown liquids. With our system, a robot holds and shakes a bottle to estimate the characteristics of the contained liquid, such as viscosity and fill level, without calculating to determine their parameters. Next, the robot pours a specified amount of the liquid into another container. The system needs to integrate information on the robot's actions, the liquids, the container, and the surrounding environment to perform the estimation and execute a continuous pouring motion using the same model. We use deep neural networks (DNN) to construct the system. The DNN model repeats prediction and execution of the actions to be taken in the next time step based on the input sensorimotor data, including camera images, force sensor data, and joint angles. At the same time, the DNN model acquires liquid characteristics in the internal state. We confirmed that the DNN model can control the robot to pour a desired amount of liquid with unknown viscosity and fill level.

    DOI

  • A Bi-directional Multiple Timescales LSTM Model for Grounding of Actions and Verbs

    Alexandre Antunes, Alban Laflaquiere, Tetsuya Ogata, Angelo Cangelosi

    IEEE International Conference on Intelligent Robots and Systems     2614 - 2621  2019年11月  [査読有り]

     概要を見る

    In this paper we present a neural architecture to learn a bi-directional mapping between actions and language. We implement a Multiple Timescale Long Short-Term Memory (MT-LSTM) network comprised of 7 layers with different timescale factors, to connect actions to language without explicitly learning an intermediate representation. Instead, the model self-organizes such representations at the level of a slow-varying latent layer, linking action branch and language branch at the center. We train the model in a bi-directional way, learning how to produce a sentence from a certain action sequence input and, simultaneously, how to generate an action sequence given a sentence as input. Furthermore we show this model preserves some of the generalization behaviour of Multiple Timescale Recurrent Neural Networks (MTRNN) in generating sentences and actions that were not explicitly trained. We compare this model with a number of different baseline models, confirming the importance of both the bi-directional training and the multiple timescales architecture. Finally, the network was evaluated on motor actions performed by an iCub robot and their corresponding letter-based description. The results of these experiments are presented at the end of the paper.

    DOI

  • Learning Multiple Sensorimotor Units to Complete Compound Tasks using an RNN with Multiple Attractors

    Kei Kase, Ryoichi Nakajo, Hiroki Mori, Tetsuya Ogata

    IEEE International Conference on Intelligent Robots and Systems     4244 - 4249  2019年11月  [査読有り]

     概要を見る

    As the complexity of the robot's tasks increases, we can consider many general tasks in a compound form that consists of shorter tasks. Therefore, for robots to generate various tasks, they need to be able to execute shorter tasks in succession, appropriately to the situation. With the design principle to construct the architecture for robots to execute complex tasks compounded with multiple subtasks, this study proposes a visuomotor-control framework with the characteristics of a state machine to train shorter tasks as sensorimotor units. The design procedure of training framework consists of 4 steps: (1) segment entire task into appropriate subtasks, (2) define subtasks as states and transitions in a state machine, (3) collect subtasks data, and (4) train neural networks: (a) autoencoder to extract visual features, (b) a single recurrent neural network to generate subtasks to realize a pseud-state-machine model with a constraint in hidden values. We implemented this framework on two different robots to allow their performance of repetitive tasks with error-recovery motion, subsequently, confirming the ability of the robot to switch the sensorimotor units from visual input at the attractors of the hidden values created by the constraint.

    DOI

  • Large-scale data collection for goal-directed drawing task with self-report psychiatric symptom questionnaires via crowdsourcing

    Shingo Murata, Hikaru Yanagida, Kentaro Katahira, Shinsuke Suzuki, Tetsuya Ogata, Yuichi Yamashita

    Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics   2019-October   3859 - 3865  2019年10月  [査読有り]

     概要を見る

    Drawing is a representative human cognitive ability and may mirror cognitive characteristics including those associated with psychiatric symptoms. Therefore, analysis of drawing data collected from various populations such as healthy people and psychiatric patients may be beneficial for better understanding human cognition. However, collecting such large-scale data about the relationship between drawing and cognitive/personality traits offline-in a laboratory-is a difficult issue. To overcome this issue, we devised a novel experimental paradigm involving a goal-directed drawing task conducted online-on the eb-with participants recruited via a crowdsourcing platform. With the assistance of 1155 participants with differing levels of psychiatric symptoms, we collected a total of 194, 040 trajectory data and answers to seven different self-report psychiatric symptom questionnaires comprising 181 items. We visualized the collected trajectory data and performed an exploratory factor analysis on the correlation matrix of the psychiatric symptom questionnaire items. Our results suggest that there were associations between psychiatric symptoms represented by specific psychiatric factors and atypical behavior observed while performing the goal-directed drawing task. This indicates the efficacy of a dimensional approach to large-scale online experiments with respect to clinical psychiatry.

    DOI

  • CNN-based multichannel end-to-end speech recognition for everyday home environments

    Nelson Yalta, Shinji Watanabe, Takaaki Hori, Kazuhiro Nakadai, Tetsuya Ogata

    European Signal Processing Conference   2019-September  2019年09月  [査読有り]

     概要を見る

    Casual conversations involving multiple speakers and noises from surrounding devices are common in everyday environments, which degrades the performances of automatic speech recognition systems. These challenging characteristics of environments are the target of the CHiME-5 challenge. By employing a convolutional neural network (CNN)-based multichannel end-to-end speech recognition system, this study attempts to overcome the presents difficulties in everyday environments. The system comprises of an attention-based encoder-decoder neural network that directly generates a text as an output from a sound input. The multichannel CNN encoder, which uses residual connections and batch renormalization, is trained with augmented data, including white noise injection. The experimental results show that the word error rate is reduced by 8.5% and 0.6% absolute from a single channel end-to-end and the best baseline (LF-MMI TDNN) on the CHiME-5 corpus, respectively.

    DOI

  • CNNRNNPBを用いたOne-Shotによる模倣動作生成

    伊藤洋, 山本健次郎, 森裕紀, 尾形哲也

    日本ロボット学会第37回学術講演会 予稿集     1A3-06  2019年09月

  • 双腕ロボットに向けた再帰型神経回路モデルを用いたドラミングタスクの学習,日本ロボット学会第37回学術講演会

    中島佳昭, 加瀬敬唯, 森裕紀, Claudio Zito, Andrey Barsky, 尾形 哲也

    日本ロボット学会第37回学術講演会 予稿集     1A2-05  2019年09月

  • 深層学習を用いた視覚運動モデルの異なる入出力情報によるロボット動作生成の比較

    松本昇, 加瀬敬唯, 森裕紀, 尾形哲也

    日本ロボット学会第37回学術講演会 予稿集     1A2-03  2019年09月

  • Discontinuous Stabilizing Control of Skid-Steering Mobile Robot (SSMR)

    Fady Ibrahim, A. A. Abouelsoud, Ahmed M.R. Fath El Bab, Tetsuya Ogata

    Journal of Intelligent and Robotic Systems: Theory and Applications   95 ( 2 ) 253 - 266  2019年08月  [査読有り]

     概要を見る

    A discontinuous stabilizing control of Skid-Steering Mobile Robot (SSMR) is proposed using σ transformation introduced in Astolfi (Syst. Control Lett. 27(1), 37–45, 1996). A linear time-invariant system (LTI) is obtained which is driven by state-dependent disturbance. A linear H∞ controller is designed to reduce the effect of this disturbance. Two control transformations are carried out in order to bring the system in a form suitable for σ transformation; one for the case of SSMR orientation around 0 and π and the other around ± π/2. The resulting two controllers for the two cases are blended using fuzzy logic. The closed-loop system is simulated using Matlab environment on point stabilization from different initial conditions. Results show that the proposed controller guarantees asymptotic stability with smooth paths. Experimental results are consistent with simulation and show that the proposed controller succeeded to stabilize the SSMR to the desired point without shuttering.

    DOI

  • Looking Back and Ahead: Adaptation and Planning by Gradient Descent

    Shingo Murata, Hiroki Sawa, Shigeki Sugano, Tetsuya Ogata

    2019 Joint IEEE 9th International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2019     151 - 156  2019年08月  [査読有り]

     概要を見る

    Adaptation and planning are crucial for both biological and artificial agents. In this study, we treat these as an inference problem that we solve using a gradient-based optimization approach. We propose adaptation and planning by gradient descent (APGraDe), a gradient-based computational framework with a hierarchical recurrent neural network (RNN) for adaptation and planning. This framework computes (counterfactual) prediction errors by looking back on past situations based on actual observations and by looking ahead to future situations based on preferred observations (or goal). The internal state of the higher level of the RNN is optimized in the direction of minimizing these errors. The errors for the past contribute to the adaptation while errors for the future contribute to the planning. The proposed APGraDe framework is implemented in a humanoid robot and the robot performs a ball manipulation task with a human experimenter. Experimental results show that given a particular preference, the robot can adapt to unexpected situations while pursuing its own preference through the planning of future actions.

    DOI

  • Weakly-Supervised Deep Recurrent Neural Networks for Basic Dance Step Generation

    Nelson Yalta, Shinji Watanabe, Kazuhiro Nakadai, Tetsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks   2019-July  2019年07月  [査読有り]

     概要を見る

    Synthesizing human's movements such as dancing is a flourishing research field which has several applications in computer graphics. Recent studies have demonstrated the advantages of deep neural networks (DNNs) for achieving remarkable performance in motion and music tasks with little effort for feature pre-processing. However, applying DNNs for generating dance to a piece of music is nevertheless challenging, because of 1) DNNs need to generate large sequences while mapping the music input, 2) the DNN needs to constraint the motion beat to the music, and 3) DNNs require a considerable amount of hand-crafted data. In this study, we propose a weakly supervised deep recurrent method for real-time basic dance generation with audio power spectrum as input. The proposed model employs convolutional layers and a multilayered Long Short-Term memory (LSTM) to process the audio input. Then, another deep LSTM layer decodes the target dance sequence. Notably, this end-to-end approach has 1) an auto-conditioned decode configuration that reduces accumulation of feedback error of large dance sequence, 2) uses a contrastive cost function to regulate the mapping between the music and motion beat, and 3) trains with weak labels generated from the motion beat, reducing the amount of hand-crafted data. We evaluate the proposed network based on i) the similarities between generated and the baseline dancer motion with a cross entropy measure for large dance sequences, and ii) accurate timing between the music and motion beat with an F-measure. Experimental results revealed that, after training using a small dataset, the model generates basic dance steps with low cross entropy and maintains an F-measure score similar to that of a baseline dancer.

    DOI

  • End-to-end Learning Method for Self-Driving Cars with Trajectory Recovery Using a Path-following Function

    Tadashi Onishi, Toshiyuki Motoyoshi, Yuki Suga, Hiroki Mori, Tsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks   2019-July  2019年07月  [査読有り]

     概要を見る

    We propose an end-to-end learning method for autonomous driving systems in this article. End-to-end model estimates an appropriate motor command from raw sensory signals. End-to-end model for autonomous driving systems has recently been based on neural networks, which are popular for their good recognition ability. A common problem is how to return a car to the driving lane when the car goes off the track. In our research, we collect recovery data based on the distance from a desired track (the nearest waypoint link) during a road test with a simulator. To train the recovery behavior, instead of collecting human driving data, we use a path-following module (which means the car automatically drives on a pre-decided route using the car's current position). Our proposed method is divided into three phases. In phase 1, we collect data only using a path-following module during 100 laps of driving. In phase 2, we generate driving behavior using a neural driving module trained by the data collected in phase 1. This includes switching between the accelerator, brake and steering based on a threshold. We collect further data on the recovery behavior using the path-following module during 100 laps of driving. In phase 3, we generate driving behavior using the neural driving module trained by the data collected in phases 1 and 2. To assess the proposed method, we compared the average distance from the nearest waypoint link and the average distance traveled per lap for datasets with no recovery, for datasets with random recovery, and for datasets for the proposed method with recovery. A model based on the proposed method drove well and paid more attention to the road rather than the sky and other unrelated objects automatically for both untrained and trained courses and weather.

    DOI

  • Sensorimotor input as a language generalisation tool: a neurorobotics model for generation and generalisation of noun-verb combinations with sensorimotor inputs

    Junpei Zhong, Martin Peniak, Jun Tani, Tetsuya Ogata, Angelo Cangelosi

    Autonomous Robots   43 ( 5 ) 1271 - 1290  2019年06月  [査読有り]

     概要を見る

    The paper presents a neurorobotics cognitive model explaining the understanding and generalisation of nouns and verbs combinations when a vocal command consisting of a verb-noun sentence is provided to a humanoid robot. The dataset used for training was obtained from object manipulation tasks with a humanoid robot platform; it includes 9 motor actions and 9 objects placing placed in 6 different locations), which enables the robot to learn to handle real-world objects and actions. Based on the multiple time-scale recurrent neural networks, this study demonstrates its generalisation capability using a large data-set, with which the robot was able to generalise semantic representation of novel combinations of noun-verb sentences, and therefore produce the corresponding motor behaviours. This generalisation process is done via the grounding process: different objects are being interacted, and associated, with different motor behaviours, following a learning approach inspired by developmental language acquisition in infants. Further analyses of the learned network dynamics and representations also demonstrate how the generalisation is possible via the exploitation of this functional hierarchical recurrent network.

    DOI

  • End-to-End自動運転モデル改善のための画像認識サブタスクの設計と評価

    石晶,李志豪, 本吉俊之, 大西直, 森裕紀, 尾形哲也

    第33回人工知能学会全国大会 予稿集     1L2-J-11-01  2019年06月

  • Conditional Generative Adversarial Networks によるロボットアームの障害物回避軌道計画

    鳥島亮太, 森裕紀, 高橋城志, 岡野原大輔, 尾形哲也

    日本機械学会ロボティクスメカトロニクス講演会 予稿集     1P2-A10  2019年06月

  • モータ関節角と電流値を用いた再帰型神経回路モデルによるペグ挿入動作生成

    倉田拓実, 伊藤洋, 森裕紀, 山本健次郎, 尾形哲也

    日本機械学会ロボティクスメカトロニクス講演会 予稿集     1P2-A10  2019年06月

  • Path following algorithm for skid-steering mobile robot based on adaptive discontinuous posture control

    Fady Ibrahim, A. A. Abouelsoud, Ahmed M.R. Fath Elbab, Tetsuya Ogata

    Advanced Robotics   33 ( 9 ) 439 - 453  2019年05月  [査読有り]

     概要を見る

    The kinematic model of a skid-steering mobile robot (SSMR) is manipulated using signed polar transformation which represents a discontinuous state transformation. The influence of relative position between the instantaneous center of rotation (ICR) and SSMR center of mass is considered. Then, adaptive state feedback controller is designed and stability regions are studied. Subsequently, a point-to-point tracking algorithm is introduced to track a trajectory that is defined by a set of way-points, which is the more realistic case of dangerous exploration or landmine detection purposes. The closed-loop system is simulated using MATLAB environment and experimentally validated using a modified TURTLEBOT3 Burger. Results show that the proposed controller reaches almost zero steady state error with smooth paths for point stabilization, moreover, good tracking capabilities are demonstrated. The proposed control system integrates both posture and tracking algorithm, thus achieve trajectory tacking which is defined by a set of way-points.

    DOI

  • Morphology-specific convolutional neural networks for tactile object recognition with a multi-fingered hand

    Satoshi Funabashi, Gang Yan, Andreas Geier, Alexander Schmitz, Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Conference on Robotics and Automation   2019-May   57 - 63  2019年05月  [査読有り]

     概要を見る

    Distributed tactile sensors on multi-fingered hands can provide high-dimensional information for grasping objects, but it is not clear how to optimally process such abundant tactile information. The current paper explores the possibility of using a morphology-specific convolutional neural network (MS-CNN). uSkin tactile sensors are mounted on an Allegro Hand, which provides 720 force measurements (15 patches of uSkin modules with 16 triaxial force sensors each) in addition to 16 joint angle measurements. Consecutive layers in the CNN get input from parts of one finger segment, one finger, and the whole hand. Since the sensors give 3D (x, y, z) vector tactile information, inputs with 3 channels (x, y and z) are used in the first layer, based on the idea of such inputs for RGB images from cameras. Overall, the layers are combined, resulting in the building of a tactile map based on the relative position of the tactile sensors on the hand. Seven different combination variations were evaluated, and an over-95% object recognition rate with 20 objects was achieved, even though only one random time instance from a repeated squeezing motion of an object in an unknown pose within the hand was used as input.

    DOI

  • Adaptive Drawing Behavior by Visuomotor Learning Using Recurrent Neural Networks

    Kazuma Sasaki, Tetsuya Ogata

    IEEE Transactions on Cognitive and Developmental Systems   11 ( 1 ) 119 - 128  2019年03月  [査読有り]

     概要を見る

    Drawing is a medium that represents an idea as drawn lines, and drawing behavior requires complex cognitive abilities to process visual and motor information. One way to understand aspects of these abilities is constructing computational models that can replicate these abilities rather than explaining the phenomena by building plausible models by a top-down manner. In this paper, we proposed a supervised learning model that can be trained using examples of visuomotor sequences from drawings made by human. Additionally, we demonstrated that the proposed model has functions of: 1) associating motions to depict the given picture image and 2) adapting to drawing behavior to complete a given part of the drawing process. This dynamical model is implemented by recurrent neural networks that have images and motion as their input and output. Through experiments that involved learning human drawing sequences, the model was able to associate appropriate motions to achieve depiction targets while adapting to a given part of the drawing process. Furthermore, we demonstrate that including visual information in the model improved performance robustness against noisy lines in the input data.

    DOI

  • Encoding Longer-term Contextual Sensorimotor Information in a Predictive Coding Model

    Junpei Zhong, Tetsuya Ogata, Angelo Cangelosi

    Proceedings of the 2018 IEEE Symposium Series on Computational Intelligence, SSCI 2018     160 - 167  2019年01月  [査読有り]

     概要を見る

    Studies suggest that the difference of the sensorimotor events can be recorded with the fast- and slower-changing neural activities in the hierarchical brain areas, in which they have bi-directional connections. The slow-changing representations attempt to predict the activities on the faster level by relaying categorized sensorimotor events. On the other hand, the incoming sensory information corrects such event-based prediction on the higher level by the novel or surprising signal. From this motivation, we propose a predictive hierarchical artificial neural network model which is implemented the differentiated temporal parameters for neural updates. Also, both the fast- and slow-changing neural activities are modulated by the active motor activities. This model is examined in the driving dataset, recorded in various events, which incorporates the image sequences and the ego-motion of the vehicle. Experiments show that the model encodes the driving scenarios on the higher-level where the neuron recorded the long-term context.

    DOI

  • Tool-Use Model Considering Tool Selection by a Robot Using Deep Learning

    Namiko Saito, Kitae Kim, Shingo Murata, Tetsuya Ogata, Shigeki Sugano

    IEEE-RAS International Conference on Humanoid Robots   2018-November   814 - 819  2019年01月  [査読有り]

     概要を見る

    We propose a tool-use model that can select tools that require neither labeling nor modeling of the environment and actions. With this model, a robot can choose a tool by itself and perform the operation that matches a human command and the environmental situation. To realize this, we use deep learning to train sensory motor data recorded during tool selection and tool use as experienced by a robot. The experience includes two types of selection, namely according to function and according to size, thereby allowing the robot to handle both situations. For evaluation, the robot is required to generate motion either in an untrained situation or using an untrained tool. We confirm that the robot can choose and use a tool that is suitable for achieving the target task.

    DOI

  • From natural to artificial embodied intelligence: is Deep Learning the solution (NII Shonan Meeting 137).

    Lorenzo Jamone, Tetsuya Ogata, Beata J. Grzyb

    NII Shonan Meet. Rep.   2019  2019年

  • Editorial: Machine learning methods for high-level cognitive capabilities in robotics

    Tadahiro Taniguchi, Emre Ugur, Tetsuya Ogata, Takayuki Nagai, Yiannis Demiris

    Frontiers in Neurorobotics   13  2019年

    DOI

  • Achieving Human–Robot Collaboration with Dynamic Goal Inference by Gradient Descent

    Shingo Murata, Wataru Masuda, Jiayi Chen, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano

    In Proceedings of the 26th International Conference on Neural Information Processing (ICONIP 2019)   11954 LNCS   579 - 590  2019年  [査読有り]

    DOI

  • Multisensory Learning Framework for Robot Drumming.

    Andrey Barsky, Claudio Zito, Hiroki Mori, Tetsuya Ogata, Jeremy L. Wyatt

    CoRR   abs/1907.09775  2019年  [査読有り]

  • 深層学習を用いた実機ロボットアームの高精度動作生成

    後藤守規, 伊藤洋, 森裕紀, 山本健次郎, 尾形哲也

    計測自動制御学会システムインテグレーション部門講演会SI2018 予稿集     3A3-07  2018年12月

  • AFA-PredNet: The Action Modulation Within Predictive Coding

    Junpei Zhong, Angelo Cangelosi, Xinzheng Zhang, Tetsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks   2018-July  2018年10月  [査読有り]

     概要を見る

    The predictive processing (PP) hypothesizes that the predictive inference of our sensorimotor system is encoded implicitly in the regularities between perception and action. We propose a neural architecture in which such regularities of active inference are encoded hierarchically. We further suggest that this encoding emerges during the embodied learning process when the appropriate action is selected to minimize the prediction error in perception. Therefore, this predictive stream in the sensorimotor loop is generated in a top-down manner. Specifically, it is constantly modulated by the motor actions and is updated by the bottom-up prediction error signals. In this way, the top-down prediction originally comes from the prior experience from both perception and action representing the higher levels of this hierarchical cognition. In our proposed embodied model, we extend the PredNet Network, a hierarchical predictive coding network, with the motor action units implemented by a multi-layer perceptron network (MLP) to modulate the network top-down prediction. Two experiments, a minimalistic world experiment, and a mobile robot experiment are conducted to evaluate the proposed model in a qualitative way. In the neural representation, it can be observed that the causal inference of predictive percept from motor actions can be also observed while the agent is interacting with the environment.

    DOI

  • End-to-End Visuomotor Learning of Drawing Sequences using Recurrent Neural Networks

    Kazuma Sasaki, Tetsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks   2018-July  2018年10月  [査読有り]

     概要を見る

    Drawing is one of the complex cognitive abilities of humans. Cognitive neuropsychological studies have attempted to develop models that can explain the observations of the drawing behavior. These models exhibit limitations to reproduce the drawing behaviors because of individual factors that are related to the drawing style or non-reproducibility of motions. A constructive approach provides another methodology to investigate the complex systems by constructing models that can reproducibly replicate the behaviors. In this study, we focus on an ability to reuse the integrated visuomotor memory of drawing to associate the drawing motion from an image. Existing computational models of drawing have not considered the visual information in hand-drawn pictures. Therefore, we propose a dynamical model of the visuomotor process of drawing. The proposed model does not require any prior knowledge of the process such as the pre-designed shape primitives or the image processing algorithms. The proposed model is implemented by utilizing a recurrent neural network that learns the visuomotor transition of the drawing process. The association of the model's drawing motion by reusing the obtained memory can be obtained by minimizing the prediction error of the image. By performing simulator experiments, the proposed model demonstrates its association ability in case of pictures that comprise multiple lines.

    DOI

  • Motion switching with sensory and instruction signals by designing dynamical systems using deep neural network

    Kanata Suzuki, Hiroki Mori, Tetsuya Ogata

    IEEE Robotics and Automation Letters   3 ( 4 ) 3481 - 3488  2018年10月  [査読有り]

     概要を見る

    To ensure that a robot is able to accomplish an extensive range of tasks, it is necessary to achieve a flexible combination of multiple behaviors. This is because the design of task motions suited to each situation would become increasingly difficult as the number of situations and the types of tasks performed by them increase. To handle the switching and combination of multiple behaviors, we propose a method to design dynamical systems based on point attractors that accept (i) 'instruction signals' for instruction-driven switching. We incorporate the (ii) 'instruction phase' to form a point attractor and divide the target task into multiple subtasks. By forming an instruction phase that consists of point attractors, the model embeds a subtask in the form of trajectory dynamics that can be manipulated using sensory and instruction signals. Our model comprises two deep neural networks: A convolutional autoencoder and a multiple time-scale recurrent neural network. In this study, we apply the proposed method to manipulate soft materials. To evaluate our model, we design a cloth-folding task that consists of four subtasks and three patterns of instruction signals, which indicate the direction of motion. The results depict that the robot can perform the required task by combining subtasks based on sensory and instruction signals. And, our model determined the relations among these signals using its internal dynamics.

    DOI

  • Paired recurrent autoencoders for bidirectional translation between robot actions and linguistic descriptions

    Tatsuro Yamada, Hiroyuki Matsunaga, Tetsuya Ogata

    IEEE Robotics and Automation Letters   3 ( 4 ) 3441 - 3448  2018年10月  [査読有り]

     概要を見る

    We propose a novel deep learning framework for bidirectional translation between robot actions and their linguistic descriptions. Our model consists of two recurrent autoencoders (RAEs). One RAE learns to encode action sequences as fixed-dimensional vectors in a way that allows the sequences to be reproduced from the vectors by its decoder. The other RAE learns to encode descriptions in a similar way. In the learning process, in addition to reproduction losses, we create another loss function whereby the representations of an action and its corresponding description approach each other in the latent vector space. Across the shared representation, the trained model can produce a linguistic description given a robot action. The model is also able to generate an appropriate action by receiving a linguistic instruction, conditioned on the current visual input. Visualization of the latent representations shows that the robot actions are embedded in a semantically compositional way in the vector space by being learned jointly with descriptions.

    DOI

  • Put-in-Box Task Generated from Multiple Discrete Tasks by aHumanoid Robot Using Deep Learning

    Kei Kase, Kanata Suzuki, Pin Chu Yang, Hiroki Mori, Tetsuya Ogata

    Proceedings - IEEE International Conference on Robotics and Automation     6447 - 6452  2018年09月  [査読有り]

     概要を見る

    For robots to have a wide range of applications, they must be able to execute numerous tasks. However, recent studies into robot manipulation using deep neural networks (DNN) have primarily focused on single tasks. Therefore, we investigate a robot manipulation model that uses DNNs and can execute long sequential dynamic tasks by performing multiple short sequential tasks at appropriate times. To generate compound tasks, we propose a model comprising two DNNs: a convolutional autoencoder that extracts image features and a multiple timescale recurrent neural network (MTRNN) to generate motion. The internal state of the MTRNN is constrained to have similar values at the initial and final motion steps; thus, motions can be differentiated based on the initial image input. As an example compound task, we demonstrate that the robot can generate a 'Put-In-Box' task that is divided into three subtasks: open the box, grasp the object and put it into the box, and close the box. The subtasks were trained as discrete tasks, and the connections between each subtask were not trained. With the proposed model, the robot could perform the Put-In-Box task by switching among subtasks and could skip or repeat subtasks depending on the situation.

    DOI

  • Message from the conference chairs

    Tetsuya Ogata, Angelo Cangelosi

    2018 Joint IEEE 8th International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2018    2018年09月

    DOI

  • Dynamic motion generation by flexible-joint robot based on deep learning using images

    Yuheng Wu, Kuniyuki Takahashi, Hiroki Yamada, Kitae Kim, Shingo Murata, Shigeki Sugano, Tetsuya Ogata

    2018 Joint IEEE 8th International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2018     169 - 174  2018年09月  [査読有り]

     概要を見る

    Robots with flexible joints have recently been attracting attention from researchers because such robots can passively adapt to environmental changes and realize dynamic motion that uses inertia. In previous research, body-model acquisition using deep learning was proposed and dynamic motion learning was achieved. However, using the end-effector position as a visual feedback signal to train a robot limits what the robot can know to only the relation between the task and itself, instead of the relation between the environment and itself. In this research, we propose to use images as a feedback signal so that the robot can have a sense of the overall situation within the task environment. This motion learning is performed via deep learning using raw image data. In an experiment, we let a robot perform task motions once to acquire motor and image data. Then, we used a convolutional auto-encoder to extract image features from raw image data. The extracted image features were used in combination with motor data to train a recurrent neural network. As a result, motion learning through deep learning from image data allowed the robot to acquire environmental information and conduct tasks that require consideration of environmental changes, making use of its advantage of passive adaptation.

    DOI

  • Detecting features of tools, objects, and actions from effects in a robot using deep learning

    Namiko Saito, Kitae Kim, Shingo Murata, Tetsuya Ogata, Shigeki Sugano

    2018 Joint IEEE 8th International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2018     91 - 96  2018年09月  [査読有り]

     概要を見る

    We propose a tool-use model that can detect the features of tools, target objects, and actions from the provided effects of object manipulation. We construct a model that enables robots to manipulate objects with tools, using infant learning as a concept. To realize this, we train sensory-motor data recorded during a tool-use task performed by a robot with deep learning. Experiments include four factors: (1) tools, (2) objects, (3) actions, and (4) effects, which the model considers simultaneously. For evaluation, the robot generates predicted images and motions given information of the effects of using unknown tools and objects. We confirm that the robot is capable of detecting features of tools, objects, and actions by learning the effects and executing the task.

    DOI

  • Learning to Achieve Different Levels of Adaptability for Human-Robot Collaboration Utilizing a Neuro-Dynamical System

    Shingo Murata, Yuxi Li, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano

    IEEE Transactions on Cognitive and Developmental Systems   10 ( 3 ) 712 - 725  2018年09月  [査読有り]

     概要を見る

    Intelligent robots are expected to collaboratively work with humans in dynamically changing daily life environments. To realize successful human-robot collaboration, robots need to deal with latent spatiotemporal complexity in the workspace and the task. To overcome this crucial issue, three levels of adaptability-motion modification, action selection, and role switching-should be considered. This paper demonstrates that a single hierarchically organized neuro-dynamical system called a multiple timescale recurrent neural network can achieve these levels of adaptability by utilizing hierarchical and bidirectional information processing. The system is implemented in a humanoid robot and the robot is required to learn to perform collaborative tasks in which some parts must be performed by a human partner and others by the robot. Experimental results show that the robot can perform collaborative tasks under dynamically changing environments, including both learned and unlearned situations, thanks to different levels of adaptability acquired in the system.

    DOI

  • A Neurorobotics Simulation of Autistic Behavior Induced by Unusual Sensory Precision

    Hayato Idei, Shingo Murata, Yiwen Chen, Yuichi Yamashita, Jun Tani, Tetsuya Ogata

    Computational Psychiatry    2018年07月  [査読有り]

  • Understanding natural language sentences with word embedding and multi-modal interaction

    Junpei Zhong, Tetsuya Ogata, Angelo Cangelosi, Chenguang Yang

    7th Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics, ICDL-EpiRob 2017   2018-January   184 - 189  2018年04月  [査読有り]

     概要を見る

    Understanding and grounding human commands with natural languages have been a fundamental requirement for service robotic applications. Although there have been several attempts toward this goal, the bottleneck still exists to store and process the corpora of natural language in an interaction system. Currently, the neural- and statistical-based (N&S) natural language processing have shown potential to solve this problem. With the availability of large data-sets nowadays, these processing methods are able to extract semantic relationships while parsing a corpus of natural language (NL) text without much human design, compared with the rule-based language processing methods. In this paper, we show that how two N&S based word embedding methods, called Word2vec and GloVe, can be used in natural language understanding as pre-training tools in a multi-modal environment. Together with two different multiple time-scale recurrent neural models, they form hybrid neural language understanding models for a robot manipulation experiment.

    DOI

  • Reduced behavioral flexibility by aberrant sensory precision in autism spectrum disorder: A neurorobotics experiment

    Hayato Idei, Shingo Murata, Yiwen Chen, Yuichi Yamashita, Jun Tani, Tetsuya Ogata

    7th Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics, ICDL-EpiRob 2017   2018-January   271 - 276  2018年04月  [査読有り]

     概要を見る

    Recently, the importance of the application of computational models utilized in cognitive neuroscience to psychiatric disorders has been recognized. This study utilizes a recurrent neural network model to test aberrant sensory precision, a normative theory of autism spectrum disorder. We particularly focus on the effects of increased and decreased sensory precision on adaptive behavior based on a prediction error minimization mechanism. To distinguish dysfunction at the behavioral and network levels, we employ a humanoid robot driven by a neural network and observe ball-playing interactions with a human experimenter. Experimental results show that behavioral rigidity characteristic of autism spectrum disorder - including stopping movement and repetitive behavior - was generated from both increased and decreased sensory precision, but through different processes at the network level. These results may provide a system-level explanation of different types of behavioral rigidity in psychiatric diseases such as compulsions and stereotypies. The results also support a system-level model for autism spectrum disorder that suggests core deficits in estimating the uncertainty of sensory evidence.

    DOI

  • RNNを備えた二台ロボット間インタラクションの複雑性解析

    澤弘樹, 山田竜郎, 村田真悟, 森裕紀, 尾形哲也, 菅野重樹

    情報処理学会第80回全国大会, 予稿集     5N-07  2018年03月

  • Effective input order of dynamics learning tree

    Chyon Hae Kim, Shohei Hama, Ryo Hirai, Kuniyuki Takahashi, Hiroki Yamada, Tetsuya Ogata, Shigeki Sugano

    Advanced Robotics   32 ( 3 ) 122 - 136  2018年02月  [査読有り]

     概要を見る

    In this paper, we discuss about the learning performance of dynamics learning tree (DLT) while mainly focusing on the implementation on robot arms. We propose an input-order-designing method for DLT. DLT has been applied to the modeling of boat, vehicle, and humanoid robot. However, the relationship between the input order and the performance of DLT has not been investigated. In the proposed method, a developer is able to design an effective input order intuitively. The proposed method was validated in the model learning tasks on a simulated robot manipulator, a real robot manipulator, and a simulated vehicle. The first/second manipulator was equipped with flexible arm/finger joints that made uncertainty around the trajectories of manipulated objects. In all of the cases, the proposed method improved the performance of DLT.

    DOI

  • CNN-based MultiChannel End-to-End Speech Recognition for everyday home environments.

    Nelson Yalta, Shinji Watanabe 0001, Takaaki Hori, Kazuhiro Nakadai, Tetsuya Ogata

    CoRR   abs/1811.02735  2018年

  • Rethinking Self-driving: Multi-task Knowledge for Better Generalization and Accident Explanation Ability.

    Zhihao Li, Toshiyuki Motoyoshi, Kazuma Sasaki, Tetsuya Ogata, Shigeki Sugano

    CoRR   abs/1809.11100  2018年

  • Detecting Features of Tools, Objects, and Actions from Effects in a Robot using Deep Learning.

    Namiko Saito, Kitae Kim, Shingo Murata, Tetsuya Ogata, Shigeki Sugano

    CoRR   abs/1809.08613  2018年

  • Weakly Supervised Deep Recurrent Neural Networks for Basic Dance Step Generation.

    Nelson Yalta, Shinji Watanabe 0001, Kazuhiro Nakadai, Tetsuya Ogata

    CoRR   abs/1807.01126  2018年

  • Encoding Longer-term Contextual Multi-modal Information in a Predictive Coding Model.

    Junpei Zhong, Tetsuya Ogata, Angelo Cangelosi

    CoRR   abs/1804.06774  2018年

  • AFA-PredNet: The action modulation within predictive coding.

    Junpei Zhong, Angelo Cangelosi, Xinzheng Zhang 0001, Tetsuya Ogata

    CoRR   abs/1804.03826  2018年

  • Deep 3D pose dictionary: 3D human pose estimation from single RGB image using deep convolutional neural network

    Reda Elbasiony, Walid Gomaa, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   11141 LNCS   310 - 320  2018年  [査読有り]

     概要を見る

    In this work, we propose a new approach for 3D human pose estimation from a single monocular RGB image based on a deep convolutional neural network (CNN). The proposed method depends on reducing the huge search space of the continuous-valued 3D human poses by discretizing and approximating these continuous poses into many discrete key-poses. These key-poses constitute more restricted search space and then can be considered as multiple-class candidates of 3D human poses. Thus, a suitable classification technique is trained using a set of 3D key-poses and their corresponding RGB images to build a model to predict the 3D pose class of an input monocular RGB image. We use deep CNN as a suitable classifier because it is proven to be the most accurate technique for RGB image classification. Our approach is proven to achieve good accuracy which is comparable to the state-of-the-art methods.

    DOI

  • Encoding Longer-Term Contextual Information with Predictive Coding and Ego-Motion

    Junpei Zhong, Angelo Cangelosi, Tetsuya Ogata, Xinzheng Zhang

    Complexity   2018   1 - 15  2018年  [査読有り]

     概要を見る

    Studies suggest that, within the hierarchical architecture, the topological higher level possibly represents the scenarios of the current sensory events with slower changing activities. They attempt to predict the neural activities on the lower level by relaying the predicted information after the scenario of the sensorimotor event has been determined. On the other hand, the incoming sensory information corrects such prediction of the events on the higher level by the fast-changing novel or surprising signal. From this point, we propose a predictive hierarchical artificial neural network model that examines this hypothesis on neurorobotic platforms. It integrates the perception and action in the predictive coding framework. Moreover, in this neural network model, there are different temporal scales of predictions existing on different levels of the hierarchical predictive coding architecture, which defines the temporal memories in recording the events occurring. Also, both the fast- and the slow-changing neural activities are modulated by the motor action. Therefore, the slow-changing neurons can be regarded as the representation of the recent scenario which the sensorimotor system has encountered. The neurorobotic experiments based on the architecture were also conducted.

    DOI

  • Four-Part Harmonization: Comparison of a Bayesian Network and a Recurrent Neural Network

    Tatsuro Yamada, Tetsuro Kitahara, Hiroaki Arie, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   11265 LNCS   213 - 225  2018年  [査読有り]

     概要を見る

    In this paper, we compare four-part harmonization produced using two different machine learning models: a Bayesian network (BN) and a recurrent neural network (RNN). Four-part harmonization is widely known as a fundamental problem in harmonization, and various methods, especially based on probabilistic models such as a hidden Markov model, a weighted finite-state transducer, and a BN, have been proposed. Recently, a method using an RNN has also been proposed. In this paper, we conducted an experiment on four-part harmonization using the same data with both a BN and RNN and investigated the differences in the results between the models. The results show that these models have different tendencies. For example, the BN’s harmonies have less dissonance but especially the bass melodies are monotonous, while the RNN’s harmonies have more dissonance but especially bass melodies are smoother.

    DOI

  • Acquisition of viewpoint transformation and action mappings via sequence to sequence imitative learning by deep neural networks

    Ryoichi Nakajo, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    Frontiers in Neurorobotics   12   46 - 46  2018年  [査読有り]

     概要を見る

    We propose an imitative learning model that allows a robot to acquire positional relations between the demonstrator and the robot, and to transform observed actions into robotic actions. Providing robots with imitative capabilities allows us to teach novel actions to them without resorting to trial-and-error approaches. Existing methods for imitative robotic learning require mathematical formulations or conversion modules to translate positional relations between demonstrators and robots. The proposed model uses two neural networks, a convolutional autoencoder (CAE) and a multiple timescale recurrent neural network (MTRNN). The CAE is trained to extract visual features from raw images captured by a camera. The MTRNN is trained to integrate sensory-motor information and to predict next states. We implement this model on a robot and conducted sequence to sequence learning that allows the robot to transform demonstrator actions into robot actions. Through training of the proposedmodel, representations of actions, manipulated objects, and positional relations are formed in the hierarchical structure of the MTRNN. After training, we confirm capability for generating unlearned imitative patterns.

    DOI

  • Dynamic motion learning for multi-DOF flexible-joint robots using active–passive motor babbling through deep learning

    Takahashi, Kuniyuki, Takahashi, Kuniyuki, Ogata, Tetsuya, Nakanishi, Jun, Cheng, Gordon, Sugano, Shigeki

    Advanced Robotics   31 ( 18 ) 1002 - 1015  2017年09月  [査読有り]

     概要を見る

    © 2017 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This paper proposes a learning strategy for robots with flexible joints having multi-degrees of freedom in order to achieve dynamic motion tasks. In spite of there being several potential benefits of flexible-joint robots such as exploitation of intrinsic dynamics and passive adaptation to environmental changes with mechanical compliance, controlling such robots is challenging because of increased complexity of their dynamics. To achieve dynamic movements, we introduce a two-phase learning framework of the body dynamics of the robot using a recurrent neural network motivated by a deep learning strategy. The proposed methodology comprises a pre-training phase with motor babbling and a fine-tuning phase with additional learning of the target tasks. In the pre-training phase, we consider active and passive exploratory motions for efficient acquisition of body dynamics. In the fine-tuning phase, the learned body dynamics are adjusted for specific tasks. We demonstrate the effectiveness of the proposed methodology in achieving dynamic tasks involving constrained movement requiring interactions with the environment on a simulated robot model and an actual PR2 robot both of which have a compliantly actuated seven degree-of-freedom arm. The results illustrate a reduction in the required number of training iterations for task learning and generalization capabilities for untrained situations.

    DOI

  • Toward abstraction from multi-modal data: Empirical studies on multiple time-scale recurrent models

    Junpei Zhong, Angelo Cangelosi, Tetsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks   2017-May   3625 - 3632  2017年06月  [査読有り]

     概要を見る

    The abstraction tasks are challenging for multi-modal sequences as they require a deeper semantic understanding and a novel text generation for the data. Although the recurrent neural networks (RNN) can be used to model the context of the time-sequences, in most cases the long-term dependencies of multi-modal data make the back-propagation through time training of RNN tend to vanish in the time domain. Recently, inspired from Multiple Time-scale Recurrent Neural Network (MTRNN) [1], an extension of Gated Recurrent Unit (GRU), called Multiple Time-scale Gated Recurrent Unit (MTGRU), has been proposed [2] to learn the long-term dependencies in natural language processing. Particularly it is also able to accomplish the abstraction task for paragraphs given that the time constants are well defined. In this paper, we compare the MTRNN and MTGRU in terms of its learning performances as well as their abstraction representation on higher level (with a slower neural activation). This was done by conducting two studies based on a smaller dataset (two-dimension time sequences from non-linear functions) and a relatively large data-set (43-dimension time sequences from iCub manipulation tasks with multi-modal data). We conclude that gated recurrent mechanisms may be necessary for learning long-term dependencies in large dimension multi-modal data-sets (e.g. learning of robot manipulation), even when natural language commands was not involved. But for smaller learning tasks with simple time-sequences, generic version of recurrent models, such as MTRNN, were sufficient to accomplish the abstraction task.

    DOI

  • Tool-body assimilation model considering grasping motion through deep learning

    Kuniyuki Takahashi, Kitae Kim, Tetsuya Ogata, Shigeki Sugano

    Robotics and Autonomous Systems   91   115 - 127  2017年05月  [査読有り]

     概要を見る

    We propose a tool-body assimilation model that considers grasping during motor babbling for using tools. A robot with tool-use skills can be useful in human–robot symbiosis because this allows the robot to expand its task performing abilities. Past studies that included tool-body assimilation approaches were mainly focused on obtaining the functions of the tools, and demonstrated the robot starting its motions with a tool pre-attached to the robot. This implies that the robot would not be able to decide whether and where to grasp the tool. In real life environments, robots would need to consider the possibilities of tool-grasping positions, and then grasp the tool. To address these issues, the robot performs motor babbling by grasping and nongrasping the tools to learn the robot's body model and tool functions. In addition, the robot grasps various parts of the tools to learn different tool functions from different grasping positions. The motion experiences are learned using deep learning. In model evaluation, the robot manipulates an object task without tools, and with several tools of different shapes. The robot generates motions after being shown the initial state and a target image, by deciding whether and where to grasp the tool. Therefore, the robot is capable of generating the correct motion and grasping decision when the initial state and a target image are provided to the robot.

    DOI

  • Repeatable Folding Task by Humanoid Robot Worker Using Deep Learning

    Pin Chu Yang, Kazuma Sasaki, Kanata Suzuki, Kei Kase, Shigeki Sugano, Tetsuya Ogata

    IEEE Robotics and Automation Letters   2 ( 2 ) 397 - 403  2017年04月  [査読有り]

     概要を見る

    We propose a practical state-of-the-art method to develop a machine-learning-based humanoid robot that can work as a production line worker. The proposed approach provides an intuitive way to collect data and exhibits the following characteristics: task performing capability, task reiteration ability, generalizability, and easy applicability. The proposed approach utilizes a real-time user interface with a monitor and provides a first-person perspective using a head-mounted display. Through this interface, teleoperation is used for collecting task operating data, especially for tasks that are difficult to be applied with a conventional method. A two-phase deep learning model is also utilized in the proposed approach. A deep convolutional autoencoder extracts images features and reconstructs images, and a fully connected deep time delay neural network learns the dynamics of a robot task process from the extracted image features and motion angle signals. The 'Nextage Open' humanoid robot is used as an experimental platform to evaluate the proposed model. The object folding task utilizing with 35 trained and 5 untrained sensory motor sequences for test. Testing the trained model with online generation demonstrates a 77.8% success rate for the object folding task.

    DOI

  • Learning to Perceive the World as Probabilistic or Deterministic via Interaction with Others: A Neuro-Robotics Experiment

    Shingo Murata, Yuichi Yamashita, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano, Jun Tani

    IEEE Transactions on Neural Networks and Learning Systems   28 ( 4 ) 830 - 848  2017年04月  [査読有り]

     概要を見る

    We suggest that different behavior generation schemes, such as sensory reflex behavior and intentional proactive behavior, can be developed by a newly proposed dynamic neural network model, named stochastic multiple timescale recurrent neural network (S-MTRNN). The model learns to predict subsequent sensory inputs, generating both their means and their uncertainty levels in terms of variance (or inverse precision) by utilizing its multiple timescale property. This model was employed in robotics learning experiments in which one robot controlled by the S-MTRNN was required to interact with another robot under the condition of uncertainty about the other's behavior. The experimental results show that self-organized and sensory reflex behavior-based on probabilistic prediction-emerges when learning proceeds without a precise specification of initial conditions. In contrast, intentional proactive behavior with deterministic predictions emerges when precise initial conditions are available. The results also showed that, in situations where unanticipated behavior of the other robot was perceived, the behavioral context was revised adequately by adaptation of the internal neural dynamics to respond to sensory inputs during sensory reflex behavior generation. On the other hand, during intentional proactive behavior generation, an error regression scheme by which the internal neural activity was modified in the direction of minimizing prediction errors was needed for adequately revising the behavioral context. These results indicate that two different ways of treating uncertainty about perceptual events in learning, namely, probabilistic modeling and deterministic modeling, contribute to the development of different dynamic neuronal structures governing the two types of behavior generation schemes.

    DOI PubMed

  • Emergence of interactive behaviors between two robots by prediction error minimization mechanism

    Yiwen Chen, Shingo Murata, Hiroaki Arie, Tetsuya Ogata, Jun Tani, Shigeki Sugano

    2016 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2016     302 - 307  2017年02月  [査読有り]

     概要を見る

    This study demonstrates that the prediction error minimization (PEM) mechanism can account for the emergence of reciprocal interaction between two cognitive agents. During interactive processes, alternation of forming and deforming interactions may be triggered by various internal and external causes. We focus in particular on external causes derived from a dynamic and uncertain environment. Two small humanoid robots controlled by an identical dynamic neural network model using the PEM mechanism were trained to achieve a set of coherent ball-playing interactions between them. The two robots predict each other in a top-down way while they try to minimize the prediction errors derived from the unstable ball dynamics or the external cause in a bottom-up way by using the PEM mechanism. The experimental results showed that switching among the set of trained interactive ball plays between the two robots appears spontaneously. The analysis clarified how each complementary behavior can be generated via mutual adaptation between the two robots by undertaking top-down and bottom-up interaction in each individual dynamic neural network model by using the PEM mechanism.

    DOI

  • An effective visual programming tool for learning and using robotics middleware

    Nishimura Yumi, Suga Yuki, Ogata Tetsuya

    SII 2016 - 2016 IEEE/SICE International Symposium on System Integration     156 - 161  2017年02月  [査読有り]

     概要を見る

    To encourage further use of robotics middleware, thereby, increasing the efficiency of the development process, we propose a Web-based visual programming tool for beginners. We first suggest the following requirements: (1) to utilize robotics middleware; (2) to easily build a development environment; (3) to combine variables and commands to operate various types of robots in a straightforward manner; and (4) to easily reference both objects in visual programming and source code. Consequently, we compare our work with existing visual programming tools for programming education and for robot control. Finally, we confirm the effectiveness of our tool. We detail our programming tool for learning and using Robotics Technology (RT) middleware. In particular, we use HTML5, CSS, and JavaScript to implement a Web-based GUI that automatically generates RT-Component source code. This code is then transferred through the XML-RPC process and launched on the server machine.

    DOI

  • A reusability-based hierarchical fault-detection architecture for robot middleware and its implementation in an autonomous mobile robot system

    Tao Asato, Yuki Suga, Tetsuya Ogata

    SII 2016 - 2016 IEEE/SICE International Symposium on System Integration     150 - 155  2017年02月  [査読有り]

     概要を見る

    In this study, we proposed a fault-detection architecture for robot middleware systems and implemented it in an autonomous mobile robot system. Although robot middleware has been employed in robot development, the design methodology for fault-tolerant systems remains underdeveloped. Moreover, conventional approaches have two drawbacks. First, previous fault-detection approaches were difficult to apply to robot systems in general, because they were designed for a specific robot or task. Second, although many fault-detection approaches have been domain-independent, the fault information made available was very basic to allow debugging or fault recovery. Our goal is to develop a domain-independent framework that could deliver domain-specific information. The return value of Remote Procedure Call (RPC) can notify detailed information; however, designers are reluctant to apply it as it degrades component reusability. To address this problem, we proposed a hierarchical fault-detection architecture that layers the system on the basis of the reusability of different components. RPC communication is then applied only between component with low reusability. This minimizes the negative effects of RPC, allowing it to be used for fault-detection in various robots and tasks. To demonstrate the practicality this approach, we developed a fault-detection architecture, adapted it for use with an autonomous mobile robot, and implemented it in RT-Middleware. We demonstrated that layering the system by the frequency of reuse of the different components successfully mitigated the tendency of RPC to degrade their reusability while making detailed fault information available.

    DOI

  • Analysis of imitative interactions between humans and a robot with a neuro-dynamical system

    Shingo Murata, Kai Hirano, Hiroaki Arie, Shigeki Sugano, Tetsuya Ogata

    SII 2016 - 2016 IEEE/SICE International Symposium on System Integration     343 - 348  2017年02月  [査読有り]

     概要を見る

    Human communicative behavior is both dynamic and bidirectional. This study aims to analyze such behavior by conducting imitative interactions between human subjects and a humanoid robot that has a neuro-dynamical system. For this purpose, we take a robot-centered approach in which the change in robot performance according to difference in human partner is analyzed, rather than adopting the typical human-centered approach. A small humanoid robot equipped with a neuro-dynamical system learns imitative arm movement patterns and interacts with humans after the learning process. We analyze the interactive phenomena by different methods, including principal component analysis and use of a recurrence plot. Through this analysis, we demonstrate that different classes of interactions can be observed in the contextual dynamics of the neuro-dynamical system.

    DOI

  • Sound source localization using deep learning models

    Nelson Yalta, Kazuhiro Nakadai, Tetsuya Ogata

    Journal of Robotics and Mechatronics   29 ( 1 ) 37 - 48  2017年02月  [査読有り]

     概要を見る

    This study proposes the use of a deep neural network to localize a sound source using an array of microphones in a reverberant environment. During the last few years, applications based on deep neural networks have performed various tasks such as image classification or speech recognition to levels that exceed even human capabilities. In our study, we employ deep residual networks, which have recently shown remarkable performance in image classification tasks even when the training period is shorter than that of other models. Deep residual networks are used to process audio input similar to multiple signal classification (MUSIC) methods. We show that with end-to-end training and generic preprocessing, the performance of deep residual networks not only surpasses the block level accuracy of linear models on nearly clean environments but also shows robustness to challenging conditions by exploiting the time delay on power information.

    DOI

  • ロボットミドルウェアにおける物体認識・リーチングのためのフレームワークの提案

    太田 博己, 安里 太緒, 菅 佑樹, 尾形 哲也

    ロボティクス・メカトロニクス講演会講演概要集   2017   2A2 - K11  2017年

     概要を見る

    <p>In this study, we propose a framework for object recognition and reaching motion in robot systems with robot middleware. There exist many elemental technologies for robot manipulation system such as object recognition and reaching. However, each technology is not systematically integrated for component-based development as open and common frameworks are undeveloped. In order to realize systematic integration of object recognition and reaching motion, we built a framework of robot middleware and proposed it as open framework. This framework is oriented reusable, exchangeable and extensible system and does not depend on specific robot middleware platforms. We implemented the proposed framework in a pick-and-place system with robot arm, and validate the system working.</p>

    DOI CiNii

  • ロボットシミュレーション環境構築フレームワーク「RTM-Unity Sim」の開発

    大西 直, 佐々木 一磨, 本吉 俊之, 菅 佑樹, 尾形 哲也

    ロボティクス・メカトロニクス講演会講演概要集   2017   2A2 - J09  2017年

     概要を見る

    <p>In recent years, several robot simulators and open source game engines have been developed. Open source game engines such as Unity and Unreal Engine 4 help to create the simulation environment. Therefore, we implemented a simulator environment by connecting a game engine and a robot middleware. The requirements for the simulator are follows: (1) To create a simulation environment with the high degree of freedom. (2) To support the multiple OS. In this study, we selected Unity and OpenRTM-aist. To assess the machine learning performance, the training data can be collected and the learned model can be verified in the simulator environment. As an example of the implementation of the simulation environment using this framework, we developed a simulator for autonomous driving systems.</p>

    DOI CiNii

  • MTRNNを用いたEnd to End Learningによる移動行動学習システムの開発

    日永田 佑介, 伊藤 洋, 山本 健二郎, 尾形 哲也

    ロボティクス・メカトロニクス講演会講演概要集   2017   2A2 - D11  2017年

     概要を見る

    <p>This paper presents an autonomous mobile system using E2E learning. We examined the autonomous mobile system which could move in real environment for service robot and autonomous mobile vehicles. As a result, we proposed a system using MTRNN for E2E learning method of the autonomous mobile system. In addition, we developed the mobile robot system for the experiments. We suggested the machine learning simulator which could collect enormous learning data at low cost and developed it. By the above, we developed learning system of the autonomous mobile system by E2E learning.</p>

    DOI CiNii

  • 神経回路モデルを用いた感覚不確実性の予測による状況変化に対する適応的行動生成

    増田 航, 村田 真悟, 富岡 咲希, 尾形 哲也, 菅野 重樹

    ロボティクス・メカトロニクス講演会講演概要集   2017   1P2 - N08  2017年

     概要を見る

    <p>Robots working in the real environment need to respond to necessary sensory inputs. However, if the sensory inputs are not necessary for action generation, robots need to stably generate action without being affected by unnecessary sensory inputs. To realize such adaptive action generation against situational changes, robots should automatically decide how much sensory inputs are necessary for action generation. In this research, we propose a method which automatically decides the ratio of actual and predicted sensory inputs based on predicted sensory uncertainty. As a result of robot experiments, the robot with proposed mechanism could conduct adaptive action generation against situational changes.</p>

    DOI CiNii

  • Online Motion Generation with Sensory Information and Instructions by Hierarchical RNN.

    Kanata Suzuki, Hiroki Mori, Tetsuya Ogata

    CoRR   abs/1712.05109  2017年

  • General problem solving with category theory.

    Francisco J. Arjonilla, Tetsuya Ogata

    CoRR   abs/1709.04825  2017年

  • Toward Abstraction from Multi-modal Data: Empirical Studies on Multiple Time-scale Recurrent Models.

    Junpei Zhong, Angelo Cangelosi, Tetsuya Ogata

    CoRR   abs/1702.05441  2017年

  • Learning of labeling room space for mobile robots based on visual motor experience

    Tatsuro Yamada, Saki Ito, Hiroaki Arie, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   10613 LNCS   35 - 42  2017年  [査読有り]

     概要を見る

    A model was developed to allow a mobile robot to label the areas of a typical domestic room, using raw sequential visual and motor data, no explicit information on location was provided, and no maps were constructed. The model comprised a deep autoencoder and a recurrent neural network. The model was demonstrated to (1) learn to correctly label areas of different shapes and sizes, (2) be capable of adapting to changes in room shape and rearrangement of items in the room, and (3) attribute different labels to the same area, when approached from different angles. Analysis of the internal representations of the model showed that a topological structure corresponding to the room structure was self-organized as the trajectory of the internal activations of the network.

    DOI

  • Representation learning of logic words by an RNN: From word sequences to robot actions

    Tatsuro Yamada, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    Frontiers in Neurorobotics   11 ( 70 ) 70 - 70  2017年  [査読有り]

     概要を見る

    An important characteristic of human language is compositionality. We can efficiently express a wide variety of real-world situations, events, and behaviors by compositionally constructing the meaning of a complex expression from a finite number of elements. Previous studies have analyzed how machine-learning models, particularly neural networks, can learn from experience to represent compositional relationships between language and robot actions with the aim of understanding the symbol grounding structure and achieving intelligent communicative agents. Such studies have mainly dealt with the words (nouns, adjectives, and verbs) that directly refer to real-world matters. In addition to these words, the current study deals with logic words, such as “not,” “and,” and “or” simultaneously. These words are not directly referring to the real world, but are logical operators that contribute to the construction of meaning in sentences. In human–robot communication, these words may be used often. The current study builds a recurrent neural network model with long short-term memory units and trains it to learn to translate sentences including logic words into robot actions. We investigate what kind of compositional representations, which mediate sentences and robot actions, emerge as the network’s internal states via the learning process. Analysis after learning shows that referential words are merged with visual information and the robot’s own current state, and the logical words are represented by the model in accordance with their functions as logical operators. Words such as “true,” “false,” and “not” work as non-linear transformations to encode orthogonal phrases into the same area in a memory cell state space. The word “and,” which required a robot to lift up both its hands, worked as if it was a universal quantifier. The word “or,” which required action generation that looked apparently random, was represented as an unstable space of the network’s dynamical system.

    DOI

  • Mixing actual and predicted sensory states based on uncertainty estimation for flexible and robust robot behavior

    Shingo Murata, Wataru Masuda, Saki Tomioka, Tetsuya Ogata, Shigeki Sugano

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   10613 LNCS   11 - 18  2017年  [査読有り]

     概要を見る

    In this paper, we propose a method to dynamically modulate the input state of recurrent neural networks (RNNs) so as to realize flexible and robust robot behavior. We employ the so-called stochastic continuous-time RNN (S-CTRNN), which can learn to predict the mean and variance (or uncertainty) of subsequent sensorimotor information. Our proposed method uses this estimated uncertainty to determine a mixture ratio for combining actual and predicted sensory states of network input. The method is evaluated by conducting a robot learning experiment in which a robot is required to perform a sensory-dependent task and a sensory-independent task. The sensory-dependent task requires the robot to incorporate meaningful sensory information, and the sensory-independent task requires the robot to ignore irrelevant sensory information. Experimental results demonstrate that a robot controlled by our proposed method exhibits flexible and robust behavior, which results from dynamic modulation of the network input on the basis of the estimated uncertainty of actual sensory states.

    DOI

  • 深層学習とマニピュレーション

    尾形 哲也

    日本ロボット学会誌   35 ( 1 ) 28 - 31  2017年  [招待有り]

    CiNii

  • Learning to Perceive the World as Probabilistic or Deterministic via Interaction With Others: A Neuro-Robotics Experiment.

    Shingo Murata, Yuichi Yamashita, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano, Jun Tani

    IEEE Trans. Neural Networks Learn. Syst.   28 ( 4 ) 830 - 848  2017年  [査読有り]

     概要を見る

    We suggest that different behavior generation schemes, such as sensory reflex behavior and intentional proactive behavior, can be developed by a newly proposed dynamic neural network model, named stochastic multiple timescale recurrent neural network (S-MTRNN). The model learns to predict subsequent sensory inputs, generating both their means and their uncertainty levels in terms of variance (or inverse precision) by utilizing its multiple timescale property. This model was employed in robotics learning experiments in which one robot controlled by the S-MTRNN was required to interact with another robot under the condition of uncertainty about the other's behavior. The experimental results show that self-organized and sensory reflex behavior-based on probabilistic prediction-emerges when learning proceeds without a precise specification of initial conditions. In contrast, intentional proactive behavior with deterministic predictions emerges when precise initial conditions are available. The results also showed that, in situations where unanticipated behavior of the other robot was perceived, the behavioral context was revised adequately by adaptation of the internal neural dynamics to respond to sensory inputs during sensory reflex behavior generation. On the other hand, during intentional proactive behavior generation, an error regression scheme by which the internal neural activity was modified in the direction of minimizing prediction errors was needed for adequately revising the behavioral context. These results indicate that two different ways of treating uncertainty about perceptual events in learning, namely, probabilistic modeling and deterministic modeling, contribute to the development of different dynamic neuronal structures governing the two types of behavior generation schemes.

    DOI

  • Visual motor integration of robot's drawing behavior using recurrent neural network

    Kazuma Sasaki, Kuniaki Noda, Tetsuya Ogata

    Robotics and Autonomous Systems   86   184 - 195  2016年12月  [査読有り]

     概要を見る

    Drawing is a way of visually expressing our feelings, knowledge, and situation. People draw pictures to share information with other human beings. This study investigates visuomotor memory (VM), which is a reusable memory storing drawing behavioral data. We propose a neural network-based model for acquiring a computational memory that can replicate VM through self-organized learning of a robot's actual drawing experiences. To design the model, we assume that VM has the following two characteristics: (1) it is formed by bottom-up learning and integration of temporal drawn pictures and motion data, and (2) it allows the observers to associate drawing motions from pictures. The proposed model comprises a deep neural network for dimensionally compressing temporal drawn images and a continuous-time recurrent neural network for integration learning of drawing motions and temporal drawn images. Two experiments are conducted on unicursal shape learning to investigate whether the proposed model can learn the function without any shape information for visual processing. Based on the first experiment, the model can learn 15 drawing sequences for three types of pictures, acquiring associative memory for drawing motions through the bottom-up learning process. Thus, it can associate drawing motions from untrained drawn images. In the second experiment, four types of pictures are trained, with four distorted variations per type. In this case, the model can organize the different shapes based on their distortions by utilizing both the image information and the drawing motions, even if visual characteristics are not shared.

    DOI

  • Achieving Different Levels of Adaptability for Human–Robot Collaboration Utilizing a Neuro-Dynamical System

    Yuxi Li, Shingo Murata, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano

    Workshop on Bio-inspired Social Robot Learning in Home Scenarios, The 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016)     1 - 6  2016年10月  [査読有り]

  • 神経回路モデルによるロボットの行動と言語の統合学習

    尾形 哲也

    計測と制御   55 ( 10 ) 872 - 877  2016年10月  [招待有り]

    CiNii

  • Symbol emergence in robotics: A survey

    Tadahiro Taniguchi, Takayuki Nagai, Tomoaki Nakamura, Naoto Iwahashi, Tetsuya Ogata, Hideki Asoh

    Advanced Robotics   30 ( 11-12 ) 706 - 728  2016年06月  [査読有り]

     概要を見る

    Humans can learn a language through physical interaction with their environment and semiotic communication with other people. It is very important to obtain a computational understanding of how humans can form symbol systems and obtain semiotic skills through their autonomous mental development. Recently, many studies have been conducted regarding the construction of robotic systems and machine learning methods that can learn a language through embodied multimodal interaction with their environment and other systems. Understanding human?-social interactions and developing a robot that can smoothly communicate with human users in the long term require an understanding of the dynamics of symbol systems. The embodied cognition and social interaction of participants gradually alter a symbol system in a constructive manner. In this paper, we introduce a field of research called symbol emergence in robotics (SER). SER represents a constructive approach towards a symbol emergence system. The symbol emergence system is socially self-organized through both semiotic communications and physical interactions with autonomous cognitive developmental agents, i.e. humans and developmental robots. In this paper, specifically, we describe some state-of-art research topics concerning SER, such as multimodal categorization, word discovery, and double articulation analysis. They enable robots to discover words and their embodied meanings from raw sensory-motor information, including visual information, haptic information, auditory information, and acoustic speech signals, in a totally unsupervised manner. Finally, we suggest future directions for research in SER.

    DOI

  • ロボティクスと深層学習(<特集>ニューラルネットワーク研究のフロンティア)

    尾形 哲也

    人工知能:人工知能学会誌   31 ( 2 ) 210 - 215  2016年03月  [招待有り]

    CiNii

  • Dynamical Integration of Language and Behavior in a Recurrent Neural Network for Human–Robot Interaction

    Tatsuro Yamada, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    Frontiers in Neurorobotics   10 ( JUL ) 5 - 5  2016年  [査読有り]

    DOI

  • Self and non-self discrimination mechanism based on predictive learning with estimation of uncertainty

    Ryoichi Nakajo, Maasa Takahashi, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9950 LNCS   228 - 235  2016年  [査読有り]

     概要を見る

    In this paper, we propose a model that can explain the mechanism of self and non-self discrimination. Infants gradually develop their abilities for self–other cognition through interaction with the environment. Predictive learning has been widely used to explain the mechanism of infants’ development. We hypothesized that infants’ cognitive abilities are developed through predictive learning and the uncertainty estimation of their sensory-motor inputs. We chose a stochastic continuous time recurrent neural network, which is a dynamical neural network model, to predict uncertainties as variances. From the perspective of cognitive developmental robotics, a predictive learning experiment with a robot was performed. The results indicate that training made the robot predict the regions related to its body more easily. We confirmed that self and non-self cognitive abilities might be acquired through predictive learning with uncertainty estimation.

    DOI

  • Classification of photo and sketch images using convolutional neural networks

    Kazuma Sasaki, Madoka Yamakawa, Kana Sekiguchi, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9887 LNCS   283 - 290  2016年  [査読有り]

     概要を見る

    Content-Based Image Retrieval (CBIR) system enables us to access images using only images as queries, instead of keywords. Photorealistic images, and hand-drawn sketch image can be used as a queries as well. Recently, convolutional neural networks (CNNs) are used to discriminate images including sketches. However, the tasks are limited to classifying only one type of images, either photo or sketch images, due to the lack of a large dataset of sketch images and the large difference of their visual characteristics. In this paper, we introduce a simple way to prepare training datasets, which can enable the CNN model to classify both types of images by color transforming photo and illustration images. Through the training experiment, we show that the proposed method contributes to the improvement of classification accuracy.

    DOI

  • Body model transition by tool grasping during motor babbling using deep learning and RNN

    Kuniyuki Takahashi, Hadi Tjandra, Tetsuya Ogata, Shigeki Sugano

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9886 LNCS   166 - 174  2016年  [査読有り]

     概要を見る

    We propose a method of tool use considering the transition process of a body model from not grasping to grasping a tool using a single model. In our previous research, we proposed a tool-body assimilation model in which a robot autonomously learns tool functions using a deep neural network (DNN) and recurrent neural network (RNN) through experiences of motor babbling. However, the robot started its motion already holding the tools. In real-life situations, the robot would make decisions regarding grasping (handling) or not grasping (manipulating) a tool. To achieve this, the robot performs motor babbling without the tool pre-attached to the hand with the same motion twice, in which the robot handles the tool or manipulates without graping it. To evaluate the model, we have the robot generate motions with showing the initial and target states. As a result, the robot could generate the correct motions with grasping decisions.

    DOI

  • Dynamical linking of positive and negative sentences to goal-oriented robot behavior by hierarchical RNN

    Tatsuro Yamada, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9886 LNCS   339 - 346  2016年  [査読有り]

     概要を見る

    Meanings of language expressions are constructed not only from words grounded in real-world matters, but also from words such as “not” that participate in the construction by working as logical operators. This study proposes a connectionist method for learning and internally representing functions that deal with both of these word groups, and grounding sentences constructed from them in corresponding behaviors just by experiencing raw sequential data of an imposed task. In the experiment, a robot implemented with a recurrent neural network is required to ground imperative positive and negative sentences given as a sequence of words in corresponding goal-oriented behavior. Analysis of the internal representations reveals that the network fulfilled the requirement by extracting XOR problems implicitly included in the target sequences and solving them by learning to represent the logical operations in its nonlinear dynamics in a self-organizing manner.

    DOI

  • Visual motor integration of robot's drawing behavior using recurrent neural network.

    Kazuma Sasaki, Kuniaki Noda, Tetsuya Ogata

    Robotics Auton. Syst.   86   184 - 195  2016年  [査読有り]

     概要を見る

    Drawing is a way of visually expressing our feelings, knowledge, and situation. People draw pictures to share information with other human beings. This study investigates visuomotor memory (VM), which is a reusable memory storing drawing behavioral data. We propose a neural network-based model for acquiring a computational memory that can replicate VM through self-organized learning of a robot's actual drawing experiences. To design the model, we assume that VM has the following two characteristics: (1) it is formed by bottom-up learning and integration of temporal drawn pictures and motion data, and (2) it allows the observers to associate drawing motions from pictures. The proposed model comprises a deep neural network for dimensionally compressing temporal drawn images and a continuous-time recurrent neural network for integration learning of drawing motions and temporal drawn images. Two experiments are conducted on unicursal shape learning to investigate whether the proposed model can learn the function without any shape information for visual processing. Based on the first experiment, the model can learn 15 drawing sequences for three types of pictures, acquiring associative memory for drawing motions through the bottom-up learning process. Thus, it can associate drawing motions from untrained drawn images. In the second experiment, four types of pictures are trained, with four distorted variations per type. In this case, the model can organize the different shapes based on their distortions by utilizing both the image information and the drawing motions, even if visual characteristics are not shared. (C) 2016 The Authors. Published by Elsevier B.V.

    DOI

  • Sound source separation for robot audition using deep learning

    Kuniaki Noda, Naoya Hashimoto, Kazuhiro Nakadai, Tetsuya Ogata

    IEEE-RAS International Conference on Humanoid Robots   2015-December   389 - 394  2015年12月  [査読有り]

     概要を見る

    Noise robust speech recognition is crucial for effective human-machine interaction in real-world environments. Sound source separation (SSS) is one of the most widely used approaches for addressing noise robust speech recognition by extracting a target speaker's speech signal while suppressing simultaneous unintended signals. However, conventional SSS algorithms, such as independent component analysis or nonlinear principal component analysis, are limited in modeling complex projections with scalability. Moreover, conventional systems required designing an independent subsystem for noise reduction (NR) in addition to the SSS. To overcome these issues, we propose a deep neural network (DNN) framework for modeling the separation function (SF) of an SSS system. By training a DNN to predict clean sound features of a target sound from corresponding multichannel deteriorated sound feature inputs, we enable the DNN to model the SF for extracting the target sound without prior knowledge regarding the acoustic properties of the surrounding environment. Moreover, the same DNN is trained to function simultaneously as a NR filter. Our proposed SSS system is evaluated using an isolated word recognition task and a large vocabulary continuous speech recognition task when either nondirectional or directional noise is accumulated in the target speech. Our evaluation results demonstrate that DNN performs noticeably better than the baseline approach, especially when directional noise is accumulated with a low signal-to-noise ratio.

    DOI

  • Effective motion learning for a flexible-joint robot using motor babbling

    Kuniyuki Takahashi, Tetsuya Ogata, Hiroki Yamada, Hadi Tjandra, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems   2015-December   2723 - 2728  2015年12月  [査読有り]

     概要を見る

    We propose a method for realizing effective dynamic motion learning in a flexible-joint robot using motor babbling. Flexible-joint robots have recently attracted attention because of their adaptiveness, safety, and, in particular, dynamic motions. It is difficult to control robots that require dynamic motion. In past studies, attractors and oscillators were designed as motion primitives of an assumed task in advance. However, it is difficult to adapt to unintended environmental changes using such methods. To overcome this problem, we use a recurrent neural network (RNN) that does not require predetermined parameters. In this research, we propose a method for facilitating effective learning. First, a robot learns simple motions via motor babbling, acquiring body dynamics using a recurrent neural network (RNN). Motor babbling is the process of movement that infants use to acquire their own body dynamics during their early days. Next, the robot learns additional motions required for a target task using the acquired body dynamics. For acquiring these body dynamics, the robot uses motor babbling with its redundant flexible joints to learn motion primitives. This redundancy implies that there are numerous possible motion patterns. In comparison to a basic learning task, the motion primitives are simply modified to adjust to the task. Next, we focus on the types of motions used in motor babbling. We classify the motions into two motion types, passive motion and active motion. Passive motion involves inertia without any torque input, whereas active motion involves a torque input. The robot acquires body dynamics from the passive motion and a means of torque generation from the active motion. As a result, we demonstrate the importance of performing prior learning via motor babbling before learning a task. In addition, task learning is made more efficient by dividing the motion into two types of motor babbling patterns.

    DOI

  • Neural network based model for visual-motor integration learning of robot's drawing behavior: Association of a drawing motion from a drawn image

    Kazuma Sasaki, Hadi Tjandra, Kuniaki Noda, Kuniyuki Takahashi, Tetsuya Ogata

    IEEE International Conference on Intelligent Robots and Systems   2015-December   2736 - 2741  2015年12月  [査読有り]

     概要を見る

    In this study, we propose a neural network based model for learning a robot's drawing sequences in an unsupervised manner. We focus on the ability to learn visual-motor relationships, which can work as a reusable memory in association of drawing motion from a picture image. Assuming that a humanoid robot can draw a shape on a pen tablet, the proposed model learns drawing sequences, which comprises drawing motion and drawn picture image frames. To learn raw pixel data without any given specific features, we utilized a deep neural network for compressing large dimensional picture images and a continuous time recurrent neural network for integration of motion and picture images. To confirm the ability of the proposed model, we performed an experiment for learning 15 sequences comprising three types of shapes. The model successfully learns all the sequences and can associate a drawing motion from a not trained picture image and a trained picture with similar success. We also show that the proposed model self-organizes its behavior according to types shapes.

    DOI

  • Attractor representations of language-behavior structure in a recurrent neural network for human-robot interaction

    Tatsuro Yamada, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    IEEE International Conference on Intelligent Robots and Systems   2015-December   4179 - 4184  2015年12月  [査読有り]

     概要を見る

    In recent years there has been increased interest in studies that explore integrative learning of language and other modalities by using neural network models. However, for practical application to human-robot interaction, the acquired semantic structure between language and meaning has to be available immediately and repeatably whenever necessary, just as in everyday communication. As a solution to this problem, this study proposes a method in which a recurrent neural network self-organizes cyclic attractors that reflect semantic structure and represent interaction flows in its internal dynamics. To evaluate this method we design a simple task in which a human verbally directs a robot, which responds appropriately. Training the network with training data that represent the interaction series, the cyclic attractors that reflect the semantic structure is self-organized. The network first receives a verbal direction, and its internal state moves according to the first half of the cyclic attractors with branch structures corresponding to semantics. After that, the internal state reaches a potential to generate appropriate behavior. Finally, the internal state moves to the second half and converges on the initial point of the cycle while generating the appropriate behavior. By self-organizing such an internal structure in its forward dynamics, the model achieves immediate and repeatable response to linguistic directions. Furthermore, the network self-organizes a fixed-point attractor, and so able to wait for directions. It can thus repeat the interaction flexibly without explicit turn-taking signs.

    DOI

  • Acquisition of viewpoint representation in imitative learning from own sensory-motor experiences

    Ryoichi Nakajo, Shingo Murata, Hiroaki Arie, Tetsuya Ogata

    5th Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2015     326 - 331  2015年12月  [査読有り]

     概要を見る

    This paper introduces an imitative model that enables a robot to acquire viewpoints of the self and others from its own sensory-motor experiences. This is important for recognizing and imitating actions generated from various directions. Existing methods require coordinate transformations input by human designers or complex learning modules to acquire a viewpoint. In the proposed model, several neurons dedicated to generated actions and viewpoints of the self and others are added to a dynamic nueral network model reffered as continuous time recurrent neural network (CTRNN). The training data are labeled with types of actions and viewpoints, and are linked to each internal state. We implemented this model in a robot and trained the model to perform actions of object manipulation. Representations of behavior and viewpoint were formed in the internal states of the CTRNN. In addition, we analyzed the initial values of the internal states that represent the viewpoint information. We confirmed the distinction of the observational perspective of other's actions self-organized in the space of the initial values. Combining the initial values of the internal states that describe the behavior and the viewpoint, the system can generate unlearned data.

    DOI

  • Predictive learning with uncertainty estimation for modeling infants' cognitive development with caregivers: A neurorobotics experiment

    Shingo Murata, Saki Tomioka, Ryoichi Nakajo, Tatsuro Yamada, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano

    5th Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL-EpiRob 2015     302 - 307  2015年12月  [査読有り]

     概要を見る

    Dynamic interactions with caregivers are essential for infants to develop cognitive abilities, including aspects of action, perception, and attention. We hypothesized that these abilities can be acquired through the predictive learning of sensory inputs including their uncertainty (inverse precision) in terms of variance. To examine our hypothesis from the perspective of cognitive developmental robotics, we conducted a neurorobotics experiment involving a ball-playing interaction task between a human experimenter representing a caregiver and a small humanoid robot representing an infant. The robot was equipped with a dynamic generative model called a stochastic continuous-time recurrent neural network (S-CTRNN). The S-CTRNN learned to generate predictions about both the visuo-proprioceptive states of the robot and the uncertainty of these states by minimizing a negative log-likelihood consisting of log-uncertainty and precision-weighted prediction error. The experimental results showed that predictive learning with uncertainty estimation enabled the robot to acquire infant-like cognitive abilities through dynamic interactions with the experimenter. We also discuss the effects of infant-directed modifications observed in caregiver-infant interactions on the development of these abilities.

    DOI

  • Audio-visual speech recognition using deep learning

    Kuniaki Noda, Yuki Yamaguchi, Kazuhiro Nakadai, Hiroshi G. Okuno, Tetsuya Ogata

    Applied Intelligence   42 ( 4 ) 722 - 737  2015年06月  [査読有り]

     概要を見る

    Audio-visual speech recognition (AVSR) system is thought to be one of the most promising solutions for reliable speech recognition, particularly when the audio is corrupted by noise. However, cautious selection of sensory features is crucial for attaining high recognition performance. In the machine-learning community, deep learning approaches have recently attracted increasing attention because deep neural networks can effectively extract robust latent features that enable various recognition algorithms to demonstrate revolutionary generalization capabilities under diverse application conditions. This study introduces a connectionist-hidden Markov model (HMM) system for noise-robust AVSR. First, a deep denoising autoencoder is utilized for acquiring noise-robust audio features. By preparing the training data for the network with pairs of consecutive multiple steps of deteriorated audio features and the corresponding clean features, the network is trained to output denoised audio features from the corresponding features deteriorated by noise. Second, a convolutional neural network (CNN) is utilized to extract visual features from raw mouth area images. By preparing the training data for the CNN as pairs of raw images and the corresponding phoneme label outputs, the network is trained to predict phoneme labels from the corresponding mouth area input images. Finally, a multi-stream HMM (MSHMM) is applied for integrating the acquired audio and visual HMMs independently trained with the respective features. By comparing the cases when normal and denoised mel-frequency cepstral coefficients (MFCCs) are utilized as audio features to the HMM, our unimodal isolated word recognition results demonstrate that approximately 65 % word recognition rate gain is attained with denoised MFCCs under 10 dB signal-to-noise-ratio (SNR) for the audio signal input. Moreover, our multimodal isolated word recognition results utilizing MSHMM with denoised MFCCs and acquired visual features demonstrate that an additional word recognition rate gain is attained for the SNR conditions below 10 dB.

    DOI

  • Preferential training of neurodynamical model based on predictability of target dynamics

    Shun Nishide, Harumitsu Nobuta, Hiroshi G. Okuno, Tetsuya Ogata

    Advanced Robotics   29 ( 9 ) 587 - 596  2015年05月  [査読有り]

     概要を見る

    Intrinsic motivation is one of the keys in implementing the mechanism of interest to robots. In this paper, we present a method to apply intrinsic motivation in dynamics learning with predictable and unpredictable targets in view. The robots arm is used for the predictable target and the humans arm is used for the unpredictable target in the experiment. The learning algorithm based on intrinsic motivation will automatically set a larger weight to targets that would contribute to decreasing the training error, while setting a smaller weight to others. A neurodynamical model, namely multiple timescales recurrent neural network (MTRNN), is utilized for studying the robots arm/external object dynamics. Training of MTRNN is done using the back propagation through time (BPTT) algorithm. We modify the BPTT algorithm by the following two steps. (1) Evaluate predictability of robots arm/objects using training error of MTRNN. (2) Assign a preference ratio, which represents the weight of the training, to each object based on predictability. The proposed training method would focus more on reducing training error of predictable objects compared to normal BPTT, where training error is equally treated for every object. Experiments were conducted using an actual robot platform, moving the robots arm while a human moves his arm in the robots camera view. The results of the experiment showed that the proposed training method could achieve smaller training error of the robots arm visuomotor dynamics, which is predictable from the robots motor command, compared to general training with BPTT. Evaluation of MTRNN as a forward model to predict untrained data showed that the proposed model is capable of predicting the robots hand motion, specifically with larger number of nodes.

    DOI

  • Tool-body assimilation model based on body babbling and neurodynamical system

    Kuniyuki Takahashi, Tetsuya Ogata, Hadi Tjandra, Yuki Yamaguchi, Shigeki Sugano

    Mathematical Problems in Engineering   2015  2015年02月  [査読有り]

     概要を見る

    We propose the new method of tool use with a tool-body assimilation model based on body babbling and a neurodynamical system for robots to use tools. Almost all existing studies for robots to use tools require predetermined motions and tool features; the motion patterns are limited and the robots cannot use novel tools. Other studies fully search for all available parameters for novel tools, but this leads to massive amounts of calculations. To solve these problems, we took the following approach: we used a humanoid robot model to generate random motions based on human body babbling. These rich motion experiences were used to train recurrent and deep neural networks for modeling a body image. Tool features were self-organized in parametric bias, modulating the body image according to the tool in use. Finally, we designed a neural network for the robot to generate motion only from the target image. Experiments were conducted with multiple tools for manipulating a cylindrical target object. The results show that the tool-body assimilation model is capable of motion generation.

    DOI

  • Tactile object recognition using deep learning and dropout

    Alexander Schmitz, Yusuke Bansho, Kuniaki Noda, Hiroyasu Iwata, Tetsuya Ogata, Shigeki Sugano

    IEEE-RAS International Conference on Humanoid Robots   2015-February   1044 - 1050  2015年02月  [査読有り]

     概要を見る

    Recognizing grasped objects with tactile sensors is beneficial in many situations, as other sensor information like vision is not always reliable. In this paper, we aim for multimodal object recognition by power grasping of objects with an unknown orientation and position relation to the hand. Few robots have the necessary tactile sensors to reliably recognize objects: in this study the multifingered hand of TWENDY-ONE is used, which has distributed skin sensors covering most of the hand, 6 axis F/T sensors in each fingertip, and provides information about the joint angles. Moreover, the hand is compliant. When using tactile sensors, it is not clear what kinds of features are useful for object recognition. Recently, deep learning has shown promising results. Nevertheless, deep learning has rarely been used in robotics and to our best knowledge never for tactile sensing, probably because it is difficult to gather many samples with tactile sensors. Our results show a clear improvement when using a denoising autoencoder with dropout compared to traditional neural networks. Nevertheless, a higher number of layers did not prove to be beneficial.

    DOI

  • Special Issue on Cutting Edge of Robotics in Japan 2015

    Tetsuya Ogata

    Advanced Robotics   29 ( 1 ) 1 - 1  2015年01月

    DOI

  • Symbol Emergence in Robotics: A Survey.

    Tadahiro Taniguchi, Takayuki Nagai, Tomoaki Nakamura, Naoto Iwahashi, Tetsuya Ogata, Hideki Asoh

    CoRR   abs/1509.08973  2015年

  • Efficient motor babbling using variance predictions from a recurrent neural network

    Kuniyuki Takahashi, Kanata Suzuki, Tetsuya Ogata, Hadi Tjandra, Shigeki Sugano

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9491   26 - 33  2015年  [査読有り]

     概要を見る

    We propose an exploratory form of motor babbling that uses variance predictions from a recurrent neural network as a method to acquire the body dynamics of a robot with flexible joints. In conventional research methods, it is difficult to construct real robots because of the large number of motor babbling motions required. In motor babbling, different motions may be easy or difficult to predict. The variance is large in difficult-to-predict motions, whereas the variance is small in easy-topredict motions. We use a Stochastic Continuous Timescale Recurrent Neural Network to predict the accuracy and variance of motions. Using the proposed method, a robot can explore motions based on variance. To evaluate the proposed method, experiments were conducted in which the robot learns crank turning and door opening/closing tasks after exploring its body dynamics. The results show that the proposed method is capable of efficient motion generation for any given motion tasks.

    DOI

  • Audio-visual speech recognition using deep learning.

    Kuniaki Noda, Yuki Yamaguchi, Kazuhiro Nakadai, Hiroshi G. Okuno, Tetsuya Ogata

    Appl. Intell.   42 ( 4 ) 722 - 737  2015年  [査読有り]

     概要を見る

    Audio-visual speech recognition (AVSR) system is thought to be one of the most promising solutions for reliable speech recognition, particularly when the audio is corrupted by noise. However, cautious selection of sensory features is crucial for attaining high recognition performance. In the machine-learning community, deep learning approaches have recently attracted increasing attention because deep neural networks can effectively extract robust latent features that enable various recognition algorithms to demonstrate revolutionary generalization capabilities under diverse application conditions. This study introduces a connectionist-hidden Markov model (HMM) system for noise-robust AVSR. First, a deep denoising autoencoder is utilized for acquiring noise-robust audio features. By preparing the training data for the network with pairs of consecutive multiple steps of deteriorated audio features and the corresponding clean features, the network is trained to output denoised audio features from the corresponding features deteriorated by noise. Second, a convolutional neural network (CNN) is utilized to extract visual features from raw mouth area images. By preparing the training data for the CNN as pairs of raw images and the corresponding phoneme label outputs, the network is trained to predict phoneme labels from the corresponding mouth area input images. Finally, a multi-stream HMM (MSHMM) is applied for integrating the acquired audio and visual HMMs independently trained with the respective features. By comparing the cases when normal and denoised mel-frequency cepstral coefficients (MFCCs) are utilized as audio features to the HMM, our unimodal isolated word recognition results demonstrate that approximately 65 % word recognition rate gain is attained with denoised MFCCs under 10 dB signal-to-noise-ratio (SNR) for the audio signal input. Moreover, our multimodal isolated word recognition results utilizing MSHMM with denoised MFCCs and acquired visual features demonstrate that an additional word recognition rate gain is attained for the SNR conditions below 10 dB.

    DOI

  • Generation of sensory reflex behavior versus intentional proactive behavior in robot learning of cooperative interactions with others

    Shingo Murata, Yuichi Yamashita, Hiroaki Arie, Tetsuya Ogata, Jun Tani, Shigeki Sugano

    IEEE ICDL-EPIROB 2014 - 4th Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics     242 - 248  2014年12月  [査読有り]

     概要を見る

    This paper investigates the essential difference between two types of behavior generation schemes, namely, sensory reflex behavior generation and intentional proactive behavior generation, by proposing a dynamic neural network model referred to as stochastic multiple-timescale recurrent neural network (S-MTRNN). The proposed model was employed in an experiment involving robots learning to cooperate with others under the condition of potential unpredictability of the others' behaviors. The results of the learning experiment showed that sensory reflex behavior was generated by a self-organizing probabilistic prediction mechanism when the initial sensitivity characteristics in the network dynamics were not utilized in the learning process. In contrast, proactive behavior with a deterministic prediction mechanism was developed when the initial sensitivity was utilized. It was further shown that in situations where unexpected behaviors of others were observed, the behavioral context was re-situated by adaptation of the internal neural dynamics by means of simple sensory reflexes in the former case. In the latter case, the behavioral context was re-situated by error regression of the internal neural activity rather than by sensory reflex. The role of the top-down and bottom-up interactions in dealing with unexpected situations is discussed.

    DOI

  • Insertion of Pause in Drawing from Babbling for Robot’s Developmental Imitation Learning.

    Shun Nishide, Keita Mochizuki, Hiroshi G. Okuno, Tetsuya Ogata

    Proceedings of 2014 IEEE International Conference on Robots and Automation (ICRA 2014)     4785 - 4791  2014年09月  [査読有り]

    DOI

  • Learning to generate proactive and reactive behavior using a dynamic neural network model with time-varying variance prediction mechanism

    Shingo Murata, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano, Jun Tani

    Advanced Robotics   28 ( 17 ) 1189 - 1203  2014年09月  [査読有り]

     概要を見る

    This paper discusses a possible neurodynamic mechanism that enables self-organization of two basic behavioral modes, namely a proactive mode and a reactive mode, and of autonomous switching between these modes depending on the situation. In the proactive mode, actions are generated based on an internal prediction, whereas in the reactive mode actions are generated in response to sensory inputs in unpredictable situations. In order to investigate how these two behavioral modes can be self-organized and how autonomous switching between the two modes can be achieved, we conducted neurorobotics experiments by using our recently developed dynamic neural network model that has a capability to learn to predict time-varying variance of the observable variables. In a set of robot experiments under various conditions, the robot was required to imitate others movements consisting of alternating predictable and unpredictable patterns. The experimental results showed that the robot controlled by the neural network model was able to proactively imitate predictable patterns and reactively follow unpredictable patterns by autonomously switching its behavioral modes. Our analysis revealed that variance prediction mechanism can lead to self-organization of these abilities with sufficient robustness and generalization capabilities. © 2014 © 2014 Taylor & Francis and The Robotics Society of Japan.

    DOI

  • Deep neural network を用いたヒューマノイドロボットの適応的行動選択

    野田邦昭, 有江浩明, 菅佑樹, 尾形哲也

    ,GPU Technology Conference Japan    2014年07月

  • Multimodal integration learning of robot behavior using deep neural networks

    Kuniaki Noda, Hiroaki Arie, Yuki Suga, Tetsuya Ogata

    Robotics and Autonomous Systems   62 ( 6 ) 721 - 736  2014年06月  [査読有り]

     概要を見る

    For humans to accurately understand the world around them, multimodal integration is essential because it enhances perceptual precision and reduces ambiguity. Computational models replicating such human ability may contribute to the practical use of robots in daily human living environments; however, primarily because of scalability problems that conventional machine learning algorithms suffer from, sensory-motor information processing in robotic applications has typically been achieved via modal-dependent processes. In this paper, we propose a novel computational framework enabling the integration of sensory-motor time-series data and the self-organization of multimodal fused representations based on a deep learning approach. To evaluate our proposed model, we conducted two behavior-learning experiments utilizing a humanoid robot; the experiments consisted of object manipulation and bell-ringing tasks. From our experimental results, we show that large amounts of sensory-motor information, including raw RGB images, sound spectrums, and joint angles, are directly fused to generate higher-level multimodal representations. Further, we demonstrated that our proposed framework realizes the following three functions: (1) cross-modal memory retrieval utilizing the information complementation capability of the deep autoencoder; (2) noise-robust behavior recognition utilizing the generalization capability of multimodal features; and (3) multimodal causality acquisition and sensory-motor prediction based on the acquired causality. © 2014 Elsevier B.V. All rights reserved.

    DOI

  • 身体バブリングと再帰結合型神経回路モデルによる道具身体化〜深層学習による画像特徴量抽出〜

    高橋城志, 尾形哲也, Hadi Tjandra, 野田邦昭, 村田真悟, 有江浩明, 菅野重樹

    第28回人工知能学会全国大会   1I4-OS-09a-4  2014年05月

  • 予測精度の予測に基づいた能動的・受動的な適応行動の生成学習

    村田真悟, 山下祐一, 有江浩明, 尾形哲也, 谷淳, 菅野重樹

    第28回人工知能学会全国大会   2K4-OS-04a-3  2014年05月

  • Deep neural network による映像・音響・運動データの統合と共起

    野田邦昭, 有江浩明, 菅佑樹, 尾形哲也

    第28回人工知能学会全国大会   3H4-OS-24b-3  2014年05月

  • 異なる神経メカニズムによる能動的・受動的行動の選択

    村田真悟, 山下祐一, 有江浩明, 尾形哲也, 谷淳, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会   3P2-Q03  2014年05月

  • 神経回路モデルと身体バブリングによる道具身体化と道具機能の獲得

    高橋城志, 尾形哲也, TjandraHadi, 野田邦昭, 村田真悟, 有江浩明, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会   3P2-P02  2014年05月

  • Deep neural network を用いた感覚運動統合メカニズムによるヒューマノイドロボットの物体操作行動認識

    野田邦昭, 有江浩明, 菅佑樹, 尾形哲也

    日本機械学会ロボティクスメカトロニクス講演会   3P2-P03  2014年05月

  • 神経力学モデルと身体バブリングに基づく道具身体化と動作生成

    Hadi Tjandra, 高橋城志, 村田真悟, 有江浩明, 山口雄紀, 尾形哲也, 菅野重樹

    情報処理学会第76回全国大会     1S - 4  2014年03月

  • ロボットによる描画運動発達モデルと軌道の重み付き区間認識・学習を利用した精度向上

    望月敬太, 西出俊, 奥乃博, 尾形哲也

    情報処理学会第76回全国大会     3C - 5  2014年03月

  • S-CTRNNを用いた複数時系列パターンの記憶学習

    村田真悟, 有江浩明, 尾形哲也, 谷淳, 菅野重樹

    情報処理学会第76回全国大会     3C - 6  2014年03月

  • Deep Neural Networkを用いたマルチモーダル音声認識の為の特徴量学習

    山口雄紀, 野田邦昭, 中臺一博, 奥乃博, 尾形哲也

    情報処理学会第76回全国大会     5S - 3  2014年03月

  • Tool-body assimilation model using a neuro-dynamical system for acquiring representation of tool function and motion.

    Kuniyuki Takahashi, Tetsuya Ogata, Hadi Tjandra, Yuki Yamaguchi, Yuki Suga, Shigeki Sugano

    IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM   ( PM2-3 ) 1255 - 1260  2014年

     概要を見る

    In this paper, we propose a tool-body assimilation model that implements a multiple time-scales recurrent neural network (MTRNN). Our model allows a robot to acquire the representation of a tool function and the required motion without having any prior knowledge of the tool. It is composed of five modules: image feature extraction, body model, tool dynamics feature, tool recognition, and motion recognition. Self-organizing maps (SOM) are used for image feature extraction from raw images. The MTRNN is used for body model learning. Parametric bias (PB) nodes are used to learn tool dynamic features. The PB nodes are attached to the neurons of the MTRNN to modulate the body model. A hierarchical neural network (HNN) is implemented for tool and motion recognition. Experiments were conducted using OpenHRP3, a robotics simulator, with multiple tools. The results show that the tool-body assimilation model is capable of recognizing tools, including those having an unlearned shape, and acquires the required motions accordingly. © 2014 IEEE.

    DOI

  • Applying intrinsic motivation for visuomotor learning of robot arm motion

    Shun Nishide, Harumitsu Nobuta, Hiroshi G. Okuno, Tetsuya Ogata

    2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence, URAI 2014     364 - 367  2014年  [査読有り]

     概要を見る

    In this paper, we present a method to apply intrinsic motivation for improving visuomotor learning of robot's arm with external object in view. Multiple Timescales Recurrent Neural Network (MTRNN) is utilized for learning the robot arm/external object dynamics. Training of MTRNN is done using the Back Propagation Through Time (BPTT) algorithm. BPTT algorithm is modified as follows. 1. Evaluate predictability of robot arm/objects using training error of MTRNN. 2. Assign a preference ratio to each object based on predictability. The preference ratio represents the weight of each object to training. Experiments were conducted using an actual robot moving the arm while a human moves his arm in the robot's camera view. The result of the experiment showed that the proposed method presents better training result of robot arm visuomotor dynamics compared to general training with BPTT.

    DOI

  • Lipreading using convolutional neural network

    Kuniaki Noda, Yuki Yamaguchi, Kazuhiro Nakadai, Hiroshi G. Okuno, Tetsuya Ogata

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH     1149 - 1153  2014年  [査読有り]

     概要を見る

    In recent automatic speech recognition studies, deep learning architecture applications for acoustic modeling have eclipsed conventional sound features such as Mel-frequency cepstral co- efficients. However, for visual speech recognition (VSR) stud- ies, handcrafted visual feature extraction mechanisms are still widely utilized. In this paper, we propose to apply a convo- lutional neural network (CNN) as a visual feature extraction mechanism for VSR. By training a CNN with images of a speaker's mouth area in combination with phoneme labels, the CNN acquires multiple convolutional filters, used to extract vi- sual features essential for recognizing phonemes. Further, by modeling the temporal dependencies of the generated phoneme label sequences, a hidden Markov model in our proposed sys- Tem recognizes multiple isolated words. Our proposed system is evaluated on an audio-visual speech dataset comprising 300 Japanese words with six different speakers. The evaluation re- sults of our isolated word recognition experiment demonstrate that the visual features acquired by the CNN significantly out- perform those acquired by conventional dimensionality com- pression approaches, including principal component analysis.

  • Tool-body assimilation model using a neuro-dynamical system for acquiring representation of tool function and motion

    Kuniyuki Takahshi, Tetsuya Ogata, Hadi Tjandra, Yuki Yamaguchi, Yuki Suga, Shigeki Sugano

    IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM   ( PM2-3 ) 1255 - 1260  2014年  [査読有り]

     概要を見る

    In this paper, we propose a tool-body assimilation model that implements a multiple time-scales recurrent neural network (MTRNN). Our model allows a robot to acquire the representation of a tool function and the required motion without having any prior knowledge of the tool. It is composed of five modules: image feature extraction, body model, tool dynamics feature, tool recognition, and motion recognition. Self-organizing maps (SOM) are used for image feature extraction from raw images. The MTRNN is used for body model learning. Parametric bias (PB) nodes are used to learn tool dynamic features. The PB nodes are attached to the neurons of the MTRNN to modulate the body model. A hierarchical neural network (HNN) is implemented for tool and motion recognition. Experiments were conducted using OpenHRP3, a robotics simulator, with multiple tools. The results show that the tool-body assimilation model is capable of recognizing tools, including those having an unlearned shape, and acquires the required motions accordingly. © 2014 IEEE.

    DOI

  • The interaction between a robot and multiple people based on spatially mapping of friendliness and motion parameters

    Tsuyoshi Tasaki, Tetsuya Ogata, Hiroshi G. Okuno

    Advanced Robotics   28 ( 1 ) 39 - 51  2014年  [査読有り]

     概要を見る

    We aim to achieve interaction between a robot and multiple people. For this, robots should localize people, select an interaction partner, and act appropriately for him/her. It is difficult to deal with all these problems using only the sensors installed into the robots. We focus on that people use a rough interaction distance among other people . We divide this interaction area into different spaces based on both the interaction distances and sensor abilities of robots. Our robots localize people roughly within this divided space. To select an interaction partner, they map friendliness holding the interaction history onto the divided space, and integrate the sensor information. Furthermore, we developed a method for appropriately changing the motions, sizes, and speeds based on the distance. Our robots regard the divided spaces as Q-Learning states, and learn the motion parameters. Our robot interacted with 27 visitors. It localized a partner with an F-value of 0.76 through integration, which is higher than that of a single sensor. A factor analysis was performed on the results from questionnaires. Exciting and Friendly were the representatives of the first and second factors, respectively. For both factors, a motion with friendliness provided higher impression scores than that without friendliness. © 2013 Taylor & Francis and The Robotics Society of Japan.

    DOI

  • Tool-body assimilation model based on body babbling and a neuro-dynamical system for motion generation

    Kuniyuki Takahashi, Tetsuya Ogata, Hadi Tjandra, Shingo Murata, Hiroaki Arie, Shigeki Sugano

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   8681 LNCS   363 - 370  2014年  [査読有り]

     概要を見る

    We propose a model for robots to use tools without predetermined parameters based on a human cognitive model. Almost all existing studies of robot using tool require predetermined motions and tool features, so the motion patterns are limited and the robots cannot use new tools. Other studies use a full search for new tools; however, this entails an enormous number of calculations. We built a model for tool use based on the phenomenon of tool-body assimilation using the following approach: We used a humanoid robot model to generate random motion, based on human body babbling. These rich motion experiences were then used to train a recurrent neural network for modeling a body image. Tool features were self-organized in the parametric bias modulating the body image according to the used tool. Finally, we designed the neural network for the robot to generate motion only from the target image. © 2014 Springer International Publishing Switzerland.

    DOI

  • Learning and recognition of multiple fluctuating temporal patterns using S-CTRNN

    Shingo Murata, Hiroaki Arie, Tetsuya Ogata, Jun Tani, Shigeki Sugano

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   8681 LNCS   9 - 16  2014年  [査読有り]

     概要を見る

    In the present study, we demonstrate the learning and recognition capabilities of our recently proposed recurrent neural network (RNN) model called stochastic continuous-time RNN (S-CTRNN). S-CTRNN can learn to predict not only the mean but also the variance of the next state of the learning targets. The network parameters consisting of weights, biases, and initial states of context neurons are optimized through maximum likelihood estimation (MLE) using the gradient descent method. First, we clarify the essential difference between the learning capabilities of conventional CTRNN and S-CTRNN by analyzing the results of a numerical experiment in which multiple fluctuating temporal patterns were used as training data, where the variance of the Gaussian noise varied among the patterns. Furthermore, we also show that the trained S-CTRNN can recognize given fluctuating patterns by inferring the initial states that can reproduce the patterns through the same MLE scheme as that used for network training. © 2014 Springer International Publishing Switzerland.

    DOI

  • RTシステムの再利用性を高めるためのオープンフレームワークの開発

    菅佑樹, 尾形哲也

    計測自動制御学会 システムインテグレーション部門講演会 SI2013     3G1 - 2  2013年12月

  • 神経回路モデルを用いた道具身体化による道具機能と動作の獲得

    高橋城志, Tjandra Hadi, 山口雄紀, 菅佑樹, 菅野重樹, 尾形哲也

    計測自動制御学会 システムインテグレーション部門講演会 SI2013     2K1 - 5  2013年12月

  • マルチメディア向けグラフィカル統合開発環境「Max」とRTCを繋ぐブリッジプラグインの開発

    佐々木一磨, 寺田翔太, 有江浩明, 野田邦昭, 菅佑樹, 尾形哲也

    計測自動制御学会 システムインテグレーション部門講演会 SI2013     1B3 - 1  2013年12月

  • レコードスケッチ

    寺田翔太, 佐々木一磨, 有江浩明, 野田邦昭, 菅佑樹, 尾形哲也

    計測自動制御学会 システムインテグレーション部門講演会 SI2013     1B2 - 6  2013年12月

  • 相槌認識による聞き手の理解状態推定を利用したインタラクションのためのロボット動作制御

    田崎豪, 尾形哲也, 奥乃博

    ヒューマンインタフェース学会論文誌   15 ( 4 ) 363 - 374  2013年11月

  • Altered Prediction of Uncertainty Induced by Network Disequilibrium: A Neuro-Robotics Study

    Shingo Murata, Yuichi Yamashita, Tetsuya Ogata, Hiroaki Arie, Jun Tani, Shigeki Sugano

    Computational Psychiatry 2013, Abstract, Miami, USA    2013年10月

  • スパース再帰神経回路モデルによる人物の行動学習

    西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第31回学術講演会     2C2 - 01  2013年09月

  • 停止活動を活用した描画運動におけるロボットの発達的模倣学習

    望月敬太, 西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第31回学術講演会     1C2 - 06  2013年09月

  • Deep neural networkを用いたヒューマノイドロボットによる物体操作行動の記憶学習と行動生成

    野田邦昭, 有江浩明, 菅佑樹, 尾形哲也

    人工知能学会全国大会(第27回)   2G4-OS-19a-2  2013年06月

  • 再帰型神経回路モデルによる予測可能性を利用した自己・他者の識別

    有江浩明, 野田邦昭, 菅佑樹,谷淳, 尾形哲也

    人工知能学会全国大会(第27回)   3J3-OS-20b-1  2013年06月

  • 空間の能動的認知と身体の拡張

    尾形哲也

    第57回システム制御情報学会研究発表講演会(SCI’13)     126 - 1  2013年05月

  • RTミドルウエア利用者のためのオープンフレームワークの開発

    菅佑樹, 尾形哲也

    日本機械学会ロボティクスメカトロニクス講演会     1P1 - C03  2013年05月

  • Deep neural networkを用いた連想記憶メカニズムによるヒューマノイドロボットの適応的行動選択

    野田邦昭, 有江浩明, 菅佑樹, 尾形哲也

    日本機械学会ロボティクスメカトロニクス講演会     1P1 - B01  2013年05月

  • 再帰型神経回路モデルを用いた引き込みによる適応的な行為生成

    有江浩明, 野田邦昭, 菅佑樹,谷淳, 尾形哲也

    日本機械学会ロボティクスメカトロニクス講演会     1P1 - B03  2013年05月

  • 全探索と人間のアフォーダンスとの定量的差異の検証

    高橋城志, 尾形哲也, 岩田浩康, 菅野重樹

    第13回日本赤ちゃん学会学術集会    2013年05月

  • Robust multipitch analyzer against initialization based on latent harmonic allocation using overtone corpus

    Daichi Sakaue, Katsutoshi Itoyama, Tetsuya Ogata, Hiroshi G. Okuno

    Journal of Information Processing   21 ( 2 ) 246 - 255  2013年04月  [査読有り]

     概要を見る

    We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed "overtone corpus." Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings. © 2013 Information Processing Society of Japan.

    DOI

  • Integration of behaviors and languages with a hierarchal structure self-organized in a neuro-dynamical model

    Tetsuya OGATA, Hiroshi G. OKUNO

    IEEE Symposium Series on Computational Intelligence 2013     94 - 100  2013年04月

  • 多層神経回路モデルによる共感覚現象の学習と連想

    山口雄紀, 野田邦明, 西出俊, 奥乃博, 尾形哲也

    情報処理学会第75回全国大会    2013年03月

  • 神経回路モデルを用いたロボットの描画運動における発達的模倣学習

    望月敬太, 西出俊, 奥乃博, 尾形哲也

    情報処理学会第75回全国大会     1R - 5  2013年03月

  • 再帰結合神経回路モデルを用いた予測可能性による段階的対象選択学習

    信田春満, 西出俊, 奥乃博, 尾形哲也

    情報処理学会第75回全国大会     3R - 2  2013年03月

  • Integration of behaviors and languages with a hierarchal structure self-organized in a neuro-dynamical model

    Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 2013 IEEE Workshop on Robotic Intelligence in Informationally Structured Space, RiiSS 2013 - 2013 IEEE Symposium Series on Computational Intelligence, SSCI 2013     89 - 95  2013年  [査読有り]

     概要を見る

    This paper proposes an approach for robots to ac-quire language grounding in their robot's sensory-motor flow using neuro-dynamical models. We trained our neuro-dynamical model over a set of sentences represented as sequences of characters. For the integrated recognition, we introduced a cognitive hypothesis for integrated recognition where a human's brain separately processed the 'structure' and 'contents' of a sentence. Our model was trained with the spelling of words and their semantic role emerged in the first model. As a result of binding the model with sensory-motion patterns, we confirmed that it could associate proper word spellings with a sensory-motor flows and a semantic roles, even if an observed flow had not been learned. © 2013 IEEE.

    DOI

  • Developmental human-robot imitation learning of drawing with a neuro dynamical system

    Keita Mochizuki, Shun Nishide, Hiroshi G. Okuno, Tetsuya Ogata

    Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013     2336 - 2341  2013年  [査読有り]

     概要を見る

    This paper mainly deals with robot developmental learning on drawing and discusses the influences of physical embodiment to the task. Humans are said to develop their drawing skills through five phases: 1) Scribbling, 2) Fortuitous Realism, 3) Failed Realism, 4) Intellectual Realism, 5) Visual Realism. We implement phases 1) and 3) into the humanoid robot NAO, holding a pen, using a neuro dynamical model, namely Multiple Timescales Recurrent Neural Network (MTRNN). For phase 1), we used random arm motion of the robot as body babbling to associate motor dynamics with pen position dynamics. For phase 3), we developed incremental imitation learning to imitate and develop the robot's drawing skill using basic shapes: circle, triangle, and rectangle. We confirmed two notable features from the experiment. First, the drawing was better performed for shapes requiring arm motions used in babbling. Second, performance of clockwise drawing of circle was good from beginning, which is a similar phenomenon that can be observed in human development. The results imply the capability of the model to create a developmental robot relating to human development. © 2013 IEEE.

    DOI

  • Intersensory causality modeling using deep neural networks

    Kuniaki Noda, Hiroaki Arie, Yuki Suga, Tetsuya Ogata

    Proceedings - 2013 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2013     1995 - 2000  2013年  [査読有り]

     概要を見る

    Our brain is known to enhance perceptual precision and reduce ambiguity about sensory environment by integrating multiple sources of sensory information acquired from different modalities, such as vision, auditory and somatic sensation. From an engineering perspective, building a computational model that replicates this ability to integrate multimodal information and to self-organize the causal dependency among them, represents one of the central challenges in robotics. In this study, we propose such a model based on a deep learning framework and we evaluate the proposed model by conducting a bell ring task using a small humanoid robot. Our experimental results demonstrate that (1) the cross-modal memory retrieval function of the proposed method succeeds in generating visual sequence from the corresponding sound and bell ring motion, and (2) the proposed method leads to accurate causal dependencies among the sensory-motor sequence. © 2013 IEEE.

    DOI

  • Multimodal integration learning of object manipulation behaviors using deep neural networks

    Kuniaki Noda, Hiroaki Arie, Yuki Suga, Testuya Ogata

    IEEE International Conference on Intelligent Robots and Systems     1728 - 1733  2013年  [査読有り]

     概要を見る

    This paper presents a novel computational approach for modeling and generating multiple object manipulation behaviors by a humanoid robot. The contribution of this paper is that deep learning methods are applied not only for multimodal sensor fusion but also for sensory-motor coordination. More specifically, a time-delay deep neural network is applied for modeling multiple behavior patterns represented with multi-dimensional visuomotor temporal sequences. By using the efficient training performance of Hessian-free optimization, the proposed mechanism successfully models six different object manipulation behaviors in a single network. The generalization capability of the learning mechanism enables the acquired model to perform the functions of cross-modal memory retrieval and temporal sequence prediction. The experimental results show that the motion patterns for object manipulation behaviors are successfully generated from the corresponding image sequence, and vice versa. Moreover, the temporal sequence prediction enables the robot to interactively switch multiple behaviors in accordance with changes in the displayed objects. © 2013 IEEE.

    DOI

  • Learning and association of synaesthesia phenomenon using deep neural networks

    Yuki Yamaguchi, Kuniaki Noda, Shun Nishide, Hiroshi G. Okuno, Tetsuya Ogata

    2013 IEEE/SICE International Symposium on System Integration, SII 2013     659 - 664  2013年  [査読有り]

     概要を見る

    Robots are required to process multimodal information because the information in the real world comes from various modal inputs. However, there exist only a few robots integrating multimodal information. Humans can recognize the environment effectively by cross-modal processing. We focus on modeling synaesthesia phenomenon known to be a cross-modal perception of humans. Recently, deep neural networks (DNNs) have gained more attention and successfully applied to process high-dimensional data composed not only of single modality but also of multimodal information. We introduced DNNs to construct multimodal association model which can reconstruct one modality from the other modality. Our model is composed of two DNNs: one for image compression and the other for audio-visual sequential learning. We tried to reproduce synaesthesia phenomenon by training our model with the multimodal data acquired from psychological experiment. Cross-modal association experiment showed that our model can reconstruct the same or similar images from sound as synaesthetes, those who experience synaesthesia. The analysis of middle layers of DNNs representing multimodal features implied that DNNs self-organized the difference of perception between individual synaesthetes. © 2013 IEEE.

    DOI

  • RTミドルウエアを用いたテレプレゼンスロボット用フレームワークの開発

    菅佑樹, 尾形哲也

    SI 2012    2012年12月

  • 再帰型神経回路モデルを用いた内発的動機付けによる身体モデルの優先的学習

    信田春満, 西出俊, 奥乃博, 尾形哲也

    SI 2012    2012年12月

  • Tool-Body Assimilation Model using Neuro-Dynamical System for Acquiring Representation of Tool Function

    Yuki YAMAGUCHI, Harumitsu NOBUTA, Shun NISHIDE, Hiroshi G. OKUNO, Tetsuya OGATA

    IROS2012 Workshop on Cognitive Neuroscience Robotics   ( PM2-3 ) 6  2012年10月

  • Developmental Human-Robot Imitation Learning with Phased Structuring in Neuro Dynamical System

    Keita MOCHIZUKI, Harumitsu NOBUTA, Shun NISHIDE, Hiroshi G. OKUNO, Tetsuya OGATA

    IROS2012 Workshop on Cognitive Neuroscience Robotics   ( Pos-3 ) 6  2012年10月

  • Automatic allocation of training data for speech understanding based on multiple model combinations

    Kazunori Komatani, Mikio Nakano, Masaki Katsumaru, Kotaro Funakoshi, Tetsuya Ogata, Hiroshi G. Okuno

    IEICE Transactions on Information and Systems   E95-D ( 9 ) 2298 - 2307  2012年09月  [査読有り]

     概要を見る

    The optimal way to build speech understanding modules depends on the amount of training data available. When only a small amount of training data is available, effective allocation of the data is crucial to preventing overfitting of statistical methods. We have developed a method for allocating a limited amount of training data in accordance with the amount available. Our method exploits rule-based methods for when the amount of data is small, which are included in our speech understanding framework based on multiple model combinations, i.e., multiple automatic speech recognition (ASR) modules and multiple language understanding (LU) modules, and then allocates training data preferentially to the modules that dominate the overall performance of speech understanding. Experimental evaluation showed that our allocation method consistently outperforms baseline methods that use a single ASR module and a single LU module while the amount of training data increases. Copyright © 2012 The Institute of Electronics, Information and Communication Engineers.

    DOI

  • Automated violin fingering transcription through analysis of an audio recording

    Akira Maezawa, Katsutoshi Itoyama, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Computer Music Journal   36 ( 3 ) 57 - 72  2012年09月  [査読有り]

     概要を見る

    We present a method to recuperate fingerings for a given piece of violin music in order to recreate the timbre of a given audio recording of the piece. This is achieved by first analyzing an audio signal to determine the most likely sequence of two-dimensional fingerboard locations (string number and location along the string), which recovers elements of violin fingering relevant to timbre. This sequence is then used as a constraint for finding an ergonomic sequence of finger placements that satisfies both the sequence of notated pitch and the given fingerboard-location sequence. Fingerboard-location-sequence estimation is based on estimation of a hidden Markov model, each state of which represents a particular fingerboard location and emits a Gaussian mixture model of the relative strengths of harmonics. The relative strengths of harmonics are estimated from a polyphonic mixture using score-informed source segregation, and compensates for discrepancies between observed data and training data through mean normalization. Fingering estimation is based on the modeling of a cost function for a sequence of finger placements. We tailor our model to incorporate the playing practices of the violin. We evaluate the performance of the fingerboard-location estimator with a polyphonic mixture, and with recordings of a violin whose timbral characteristics differ significantly from that of the training data. We subjectively evaluate the fingering estimator and validate the effectiveness of tailoring the fingering model towards the violin. © 2012 Massachusetts Institute of Technology.

    DOI

  • 発達的模倣学習における神経力学モデルの段階的構造化と獲得プリミティブの解析

    望月敬太, 信田春満, 西出俊 奥乃博, 尾形哲也

    日本ロボット学会第30回学術講演会     4N3 - 4  2012年09月

  • 神経力学モデルを用いたロボットの道具身体化機構

    山口雄紀, 信田春満, 西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第30回学術講演会     4N3 - 3  2012年09月

  • MTRNNを用いたロボットの物体操作における挙動表現特徴量の自己組織化

    西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第30回学術講演会     3N1 - 7  2012年09月

  • 人とロボットの合奏のための多人数合奏の主導権推定

    水本武志, 尾形哲也, 奥乃博

    日本ロボット学会第30回学術講演会     3D2 - 3  2012年09月

  • 神経力学モデルによる自己身体領域抽出と視覚運動系の自己組織化

    信田春満, 河本献太, 野田邦昭, 佐部浩太郎, 西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第30回学術講演会     2H3 - 2  2012年09月

  • OpenRTM-aist 体験用開発ツール「RT System Builder」の開発

    菅佑樹, 尾形哲也

    日本ロボット学会第30回学術講演会     2B1 - 7  2012年09月

  • 楽曲印象軌跡に基づく楽曲検索システムの実装と評価

    西川直毅, 糸山克寿, 藤原弘将, 後藤真孝, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     1S - 7, 6  2012年03月

  • 神経回路モデルを用いた道具身体化モデルによる道具機能表現の獲得

    山口雄紀, 信田春満, 西出俊, 奥乃博, 尾形哲也

    情報処理学会第74回全国大会     4Q - 3, 7  2012年03月

  • Kinectによる楽器マスキングを用いた視聴覚統合ビートトラッキング

    糸原達彦, 水本武志, 大塚琢馬, 中臺一博, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     5ZD  2012年03月

  • 段階的に構造化する神経回路モデルを用いたロボットと人間の発達的インタラクション

    望月敬太, 信田春満, 西出俊 奥乃博, 尾形哲也

    情報処理学会第74回全国大会     5ZA  2012年03月

  • The DESIRE Model: Cross-modal emotion analysis and expression for robots

    Angelica Lim, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     5ZA  2012年03月

  • 複数音源下での擬音語による音源選択システム,

    山村祐介, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     4U  2012年03月

  • パーティクルフィルタを用いた動的環境下の複数音源追跡

    黄楊暘, 大塚琢馬, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     4U  2012年03月

  • Complex Infinite Sparse Factor Analysisによる周波数領域での音声信号のブラインド音源分離

    柳楽浩平, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     4U  2012年03月

  • 倍音コーパスを用いた初期値依存性の低い多重基本周波数推定法

    阪上大地, 糸山克寿, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     4S - 7, 7  2012年03月

  • 押弦制約付きギター演奏自動採譜システム

    矢澤一樹, 阪上大地, 糸山克寿, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     4S - 6, 7  2012年03月

  • 柔軟関節をもつ人間型ロボットにおける神経力学モデルを用いたダイナミック動作の模倣

    壷内将之, 尾形哲也, 奥乃博, 西出俊, 信田春満

    情報処理学会第74回全国大会     5P - 7, 8  2012年03月

  • 再帰型神経回路モデルを用いた視野変化予測と場所知覚ニューロンの発現

    信田春満, 河本献太, 野田邦昭, 佐部浩太郎, 奥乃博, 尾形哲也

    情報処理学会第74回全国大会     5P - 8, 8  2012年03月

  • アクセント特徴量を用いた歌声と朗読音声の識別システム

    阿曽慎平, 齋藤毅, 後藤真孝, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     6U - 9, 8  2012年03月

  • 発話中の方言変化に頑健な方言変換システム

    平山直樹, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第74回全国大会     6U - 8, 8  2012年03月

  • 音響特徴・ベース音・和音遷移を用いた自動和音認識

    糸山克寿, 尾形哲也, 奥乃博

    第94回音楽情報科学研究会,舘山寺温泉, 情報処理学会   Vol.2012-MUS-94, pp  2012年02月

  • スペクトル変化量のピーク間隔・F0・MFCCを用いた歌声と朗読音声の自動識別システム

    阿曽慎平, 齋藤毅, 後藤真孝, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    第94回音楽情報科学研究会,舘山寺温泉, 情報処理学会   Vol.2012-MUS-94, pp  2012年02月

  • Sound sources selection system by using onomatopoeic querries from multiple sound sources

    Yusuke Yamamura, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     2364 - 2369  2012年  [査読有り]

     概要を見る

    Our motivation is to develop a robot that treats auditory information in real environment because auditory information is useful for animated communications or understanding our surroundings. Interactions by using sound information need an aquisition of it and a proper sound source reference between a user and a robot leads to it. Such sound source reference is difficult due to multiple sound sources generating in real environemnt, and we use onomatopoeic representations as a representation for the reference. This paper shows a system that selects a sound source specified by a user from multiple sound sources. Users use onomatopoeias in the specification, and our system separates a mixed sound and converts separated sounds into onomatopoeias for the selection. Onomatopoeais have the ambiguity that each user gives each expression to a certain sound and we create an original similarity based on Minimum Edit Distance and acoustic features for solving its problem. In experiments, our system receives a mixed sound consisting of three sounds and a user's query as inputs, and checks a count of a consistency of a sound source selected by a system and a sound source specified by a user in 100 tests. The result shows © 2012 IEEE.

    DOI

  • Tool-body assimilation of humanoid robot using a neurodynamical system

    Shun Nishide, Jun Tani, Toru Takahashi, Hiroshi G. Okuno, Tetsuya Ogata

    IEEE Transactions on Autonomous Mental Development   4 ( 2 ) 139 - 149  2012年  [査読有り]

     概要を見る

    Researches in the brain science field have uncovered the human capability to use tools as if they are part of the human bodies (known as tool-body assimilation) through trial and experience. This paper presents a method to apply a robot's active sensing experience to create the tool-body assimilation model. The model is composed of a feature extraction module, dynamics learning module, and a tool-body assimilation module. Self-organizing map (SOM) is used for the feature extraction module to extract object features from raw images. Multiple time-scales recurrent neural network (MTRNN) is used as the dynamics learning module. Parametric bias (PB) nodes are attached to the weights of MTRNN as second-order network to modulate the behavior of MTRNN based on the properties of the tool. The generalization capability of neural networks provide the model the ability to deal with unknown tools. Experiments were conducted with the humanoid robot HRP-2 using no tool, I-shaped, T-shaped, and L-shaped tools. The distribution of PB values have shown that the model has learned that the robot's dynamic properties change when holding a tool. Motion generation experiments show that the tool-body assimilation model is capable of applying to unknown tools to generate goal-oriented motions. © 2012 IEEE.

    DOI

  • Efficient blind dereverberation and echo cancellation based on independent component analysis for actual acoustic signals

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Neural Computation   24 ( 1 ) 234 - 272  2012年  [査読有り]

     概要を見る

    This letter presents a new algorithm for blind dereverberation and echo cancellation based on independent component analysis (ICA) for actual acoustic signals. We focus on frequency domain ICA (FD-ICA) because its computational cost and speed of learning convergence are sufficiently reasonable for practical applications such as hands-free speech recognition. In applying conventional FD-ICA as a preprocessing of automatic speech recognition in noisy environments, one of the most critical problems is how to copewith reverberations. To extract a clean signal from the reverberant observation, we model the separation process in the shorttime Fourier transform domain and apply the multiple input/output inverse-filtering theorem (MINT) to the FD-ICA separation model. A naive implementation of this method is computationally expensive, because its time complexity is the second order of reverberation time. Therefore, themain issue in dereverberation is to reduce the high computational cost of ICA. In this letter, wereduce the computational complexity to the linear order of the reverberation time by using two techniques: (1) a separation model based on the independence of delayed observed signals with MINT and (2) spatial sphering for preprocessing. Experiments show that the computational cost grows in proportion to the linear order of the reverberation time and that ourmethod improves the word correctness of automatic speech recognition by 10 to 20 points in a RT20 =670 ms reverberant environment. © 2011 Massachusetts Institute of Technology.

    DOI PubMed

  • A musical robot that synchronizes with a coplayer using non-verbal cues

    Angelica Lim, Takeshi Mizumoto, Tetsuya Ogata, Hiroshi G. Okuno

    Advanced Robotics   26 ( 3-4 ) 363 - 381  2012年  [査読有り]

     概要を見る

    Music has long been used to strengthen bonds between humans. In our research, we develop musical coplayer robots with the hope that music may improve human-robot symbiosis as well. In this paper, we underline the importance of non-verbal, visual communication for ensemble synchronization at the start, during and end of a piece. We propose three cues for interplayer communication, and present a thereminplaying, singing robot that can detect them and adapt its play to a human flutist. Experiments with two naive flutists suggest that the system can recognize naturally occurring flutist gestures without requiring specialized user training. In addition, we show how the use of audio-visual aggregation can allow a robot to adapt to tempo changes quickly. © 2012 Koninklijke Brill NV, Leiden and The Robotics Society of Japan.

    DOI

  • A multimodal tempo and beat-tracking system based on audiovisual information from live guitar performances

    Tatsuhiko Itohara, Takuma Otsuka, Takeshi Mizumoto, Angelica Lim, Tetsuya Ogata, Hiroshi G. Okuno

    Eurasip Journal on Audio, Speech, and Music Processing   2012 ( 1 ) 6 - 6  2012年  [査読有り]

     概要を見る

    The aim of this paper is to improve beat-tracking for live guitar performances. Beat-tracking is a function to estimate musical measurements, for example musical tempo and phase. This method is critical to achieve a synchronized ensemble performance such as musical robot accompaniment. Beat-tracking of a live guitar performance has to deal with three challenges: tempo fluctuation, beat pattern complexity and environmenta noise. To cope with these problems, we devise an audiovisual integration method for beat-tracking. The auditory beat features are estimated in terms of tactus (phase) and tempo (period) by Spectro-Temporal Pattern Matching (STPM), robust against stationary noise. The visual beat features are estimated by tracking the position of the hand relative to the guitar using optical flow, mean shift and the Hough transform. Both estimated features are integrated using a particle filter to aggregate the multimodal information based on a beat location model and a hand's trajectory model. Experimental results confirm that our beat-tracking improves the F-measure by 8.9 points on average over the Murata beat-tracking method, which uses STPM and rule-based beat detection. The results also show that the system is capable of real-time processing with a suppressed number of particles while preserving the estimation accuracy. We demonstrate an ensemble with the humanoid HRP-2 that plays the theremin with a human guitarist. © 2012 Itohara et al; licensee Springer.

    DOI

  • Towards expressive musical robots: A cross-modal framework for emotional gesture, voice and music

    Angelica Lim, Tetsuya Ogata, Hiroshi G. Okuno

    Eurasip Journal on Audio, Speech, and Music Processing   2012 ( 1 ) 3 - 3  2012年  [査読有り]

     概要を見る

    It has been long speculated that expression of emotions from different modalities have the same underlying 'code', whether it be a dance step, musical phrase, or tone of voice. This is the first attempt to implement this theory across three modalities, inspired by the polyvalence and repeatability of robotics. We propose a unifying framework to generate emotions across voice, gesture, and music, by representing emotional states as a 4-parameter tuple of speed, intensity, regularity, and extent (SIRE). Our results show that a simple 4-tuple can capture four emotions recognizable at greater than chance across gesture and voice, and at least two emotions across all three modalities. An application for multi-modal, expressive music robots is discussed. © 2012 Lim et al; licensee Springer.

    DOI

  • A GMM sound source model for blind speech separation in under-determined conditions

    Yasuharu Hirasawa, Naoki Yasuraoka, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   7191 LNCS   446 - 453  2012年

     概要を見る

    This paper focuses on blind speech separation in under-determined conditions, that is, in the case when there are more sound sources than microphones. We introduce a sound source model based on the Gaussian mixture model (GMM) to represent a speech signal in the time-frequency domain, and derive rules for updating the model parameters using the auxiliary function method. Our GMM sound source model consists of two kinds of Gaussians: sharp ones representing harmonic parts and smooth ones representing nonharmonic parts. Experimental results reveal that our method outperforms the method based on non-negative matrix factorization (NMF) by 0.7dB in the signal-to-distortion ratio (SDR), and by 1.7dB in the signal-to-interference ratio (SIR). This means that our method effectively removes interference coming from other talkers. © 2012 Springer-Verlag.

    DOI

  • Complex extension of infinite sparse factor analysis for blind speech separation

    Kohei Nagira, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   7191 LNCS   388 - 396  2012年

     概要を見る

    We present a method of blind source separation (BSS) for speech signals using a complex extension of infinite sparse factor analysis (ISFA) in the frequency domain. Our method is robust against delayed signals that usually occur in real environments, such as reflections, short-time reverberations, and time lags of signals arriving at microphones. ISFA is a conventional non-parametric Bayesian method of BSS, which has only been applied to time domain signals because it can only deal with real signals. Our method uses complex normal distributions to estimate source signals and mixing matrix. Experimental results indicate that our method outperforms the conventional ISFA in the average signal-to-distortion ratio (SDR). © 2012 Springer-Verlag.

    DOI

  • Initialization-robust multipitch estimation based on latent harmonic allocation using overtone corpus

    Daichi Sakaue, Katsutoshi Itoyama, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings     425 - 428  2012年  [査読有り]

     概要を見る

    We present a new method for modeling the overtone structures of musical instruments that uses an overtone corpus generated using a MIDI synthesizer. Since multipitch estimation requires a joint estimation of F0's and their overtone structures, one of the most important problems is the overtone structure modeling. Latent harmonic allocation (LHA), a promising multipitch estimation method, is difficult to use for various applications because it requires appropriate prior distributions of the overtone structures, which cannot be determined from statistical evidence. Our method uses an overtone corpus to avoid the problem of setting prior distributions and instead restricts the lower and upper bounds of each overtone weight. The bounds are determined from reference signals generated by a MIDI synthesizer. Experimental results demonstrated that the overtone structures were stably and accurately estimated for a wide variety of initial settings. © 2012 IEEE.

    DOI

  • Incremental probabilistic geometry estimation for robot scene understanding

    Louis Kenzo Cahier, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     3625 - 3630  2012年  [査読有り]

     概要を見る

    Our goal is to give mobile robots a rich representation of their environment as fast as possible. Current mapping methods such as SLAM are often sparse, and scene reconstruction methods using tilting laser scanners are relatively slow. In this paper, we outline a new method for iterative construction of a geometric mesh using streaming time-of-flight range data. Our results show that our algorithm can produce a stable representation after 6 frames, with higher accuracy than raw time-of-flight data. © 2012 IEEE.

    DOI

  • Rhythm-based adaptive localization in incomplete RFID landmark environments

    Kenri Kodaka, Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Conference on Robotics and Automation     2108 - 2114  2012年  [査読有り]

     概要を見る

    This paper proposes a novel hybrid-structured model for the adaptive localization of robots combining a stochastic localization model and a rhythmic action model, for avoiding vacant spaces of landmarks efficiently. In regularly arranged landmark environments, robots may not be able to detect any landmarks for a long time during a straight-like movement. Consequently, locally diverse and smooth movement patterns need to be generated to keep the position estimation stable. Conventional approaches aiming at the probabilistic optimization cannot rapidly generate the detailed movement pattern due to a huge computational cost; therefore a simple but diverse movement structure needs to be introduced as an alternative option. We solve this problem by combining a particle filter as the stochastic localization module and the dynamical action model generating a zig-zagging motion. The validation experiments, where virtual-line-tracing tasks are exhibited on a floor-installed RFID environment, show that introducing the proposed rhythm pattern can improve a minimum error boundary and a velocity performance for arbitrary tolerance errors can be improved by the rhythm amplitude adaptation fed back by the localization deviation. © 2012 IEEE.

    DOI

  • Adaptive pitch control for robot thereminist using unscented Kalman filter

    Takeshi Mizumoto, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Studies in Computational Intelligence   431   19 - 24  2012年  [査読有り]

     概要を見る

    We present an adaptive pitch control method for a theremin playing robot in ensemble. The problem of the theremin playing is its sensitivity to the environment. This degrades the pitch accuracy because its pitch characteristics are time varying caused by, such as a co-player motion during the ensemble. We solve this problem using a state space model of this characteristics and an unscented Kalman filter. Experimental results show that our method reduces the pitch error the EKF and block-wise update method by 90% and 77% on average, and the robot can play a musical score of 72.9 cent error on average. © Springer-Verlag Berlin Heidelberg 2012.

    DOI

  • Automatic chord recognition based on probabilistic integration of acoustic features, bass sounds, and chord transition

    Katsutoshi Itoyama, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   7345 LNAI   58 - 67  2012年

     概要を見る

    We have developed a method that identifies musical chords in polyphonic musical signals. As musical chords mainly represent the harmony of music and are related to other musical elements such as melody and rhythm, we should be able to recognize chords more effectively if this interrelationship is taken into consideration. We use bass pitches as clues for improving chord recognition. The proposed chord recognition system is constructed based on Viterbi-algorithm- based maximum a posteriori estimation that uses a posterior probability based on chord features, chord transition patterns, and bass pitch distributions. Experimental results with 150 Beatles songs that has keys and no modulation showed that the recognition rate was 73.7% on average. © 2012 Springer-Verlag.

    DOI

  • Self-organization of object features representing motion using Multiple Timescales Recurrent Neural Network

    Shun Nishide, Jun Tani, Hiroshi G. Okuno, Tetsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks     1 - 8  2012年  [査読有り]

     概要を見る

    Affordance theory suggests that humans recognize the environment based on invariants. Invariants are features that describe the environment offering behavioral information to humans. Two types of invariants exist, structural invariants and transformational invariants. In our previous paper, we developed a method that self- organizes transformational invariants, or motion features, from camera images based on robot's experiences. The model used a bi-directional technique combining a recurrent neural network for dynamics learning, namely Recurrent Neural Network with Parametric Bias (RNNPB), and a hierarchical neural network for feature extraction. The bi-directional training method developed in the previous work was effective in clustering the motion of objects, but the analysis did not give good segregation results of the self-organized features (transformational invariants) among different motion types. In this paper, we present a refined model which integrates dynamics learning and feature extraction in a single model. The refined model is comprised of Multiple Timescales Recurrent Neural Network (MTRNN), which possesses better learning capability than RNNPB. Self-organization result of four types of motions have proved the model's capability to create clusters of object motions. The analysis showed that the model extracted feature sequences with different characteristics for four object motion types. © 2012 IEEE.

    DOI

  • Body area segmentation from visual scene based on predictability of neuro-dynamical system

    Harumitsu Nobuta, Kenta Kawamoto, Kuniaki Noda, Kohtaro Sabe, Shun Nishide, Hiroshi G. Okuno, Tetsuya Ogata

    Proceedings of the International Joint Conference on Neural Networks     1 - 8  2012年  [査読有り]

     概要を見る

    We propose neural models for segmenting the area of a body from visual scene based on predictability. Neuroscience has shown that a prediction model in brain, which predicts sensory-feedback from motor command, can divide the sensory-feedback into the self-motion derived feedback and other derived feedback. The prediction model is important for prediction control of the body. Previous studies in robotics of the prediction model assumed that a robot can recognize the position of its body (e.g. its hand) and that the view contains only that body part. In our models, motor commands and visual feedback (pixel image that includes not only a hand but also object and background) are input into a neural network model and then the body area is segmented and prediction model of body is acquired. Our model contains two parts: 1) An object detection model obtains a conversion system between object positions and the pixel image. 2) A movement prediction model predicts hand-object positions from motor commands and identifies the body. We confirmed that our models can segment the body/object area based on their pixel textures and discriminate between them by using prediction error. © 2012 IEEE.

    DOI

  • Who is the leader in a multiperson ensemble? - Multiperson human-robot ensemble model with leaderness - Multiperson h

    Takeshi Mizumoto, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1413 - 1419  2012年  [査読有り]

     概要を見る

    This paper presents a state space model for a multiperson ensemble and an estimation method of the onset timings, tempos, and leaders. In a multiperson ensemble, determining one explicit leader is difficult because (1) participants' rhythms are mutually influenced and (2) they compete with each other. Most ensemble studies however assumed that one leader exists at a time and the others just follow the leader. To deal with the multiple and time-varying leaders, we define leaderness indicating the power to influence the others as the product of the tempo stability and the distance from the ensemble tempo. This definition means that a leader should have a strong desire to change the current tempo. Using the leaderness, we present a state space model of a multiperson ensemble and an unscented Kalman filter based estimation method. The model consists of the leaderness update, the ensemble tempo update, the individual tempo update, and the onset timing adaptation, each of which has a relationship to psychological results of an ensemble. We evaluate our method using simulation and human behavior. The simulation results show that our model is stable for various initial tempos and the number of participants. For the human behavior, pairs and triads of participants are asked to tap keys in synchronization with the others. The results show that the leaderness successfully indicate the dynamics of the leaders, and the onset errors are 181msec and 241msec for pairs and triads on average, respectively, which are comparable to those of humans (153msec and 227msec for pairs and triads, respectively.) © 2012 IEEE.

    DOI

  • Improvement of audio-visual score following in robot ensemble with human guitarist

    Tatsuhiko Itohara, Kazuhiro Nakadai, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE-RAS International Conference on Humanoid Robots     574 - 579  2012年  [査読有り]

     概要を見る

    Our goal is to create an ensemble between human guitarists and music robots, e.g., singing and playing instruments robots. Such robots need to detect the tempo and beat time of the music. Score following and beat tracking, which requires and does not requires a score, are commonly used for this purpose. Score following is an incremental audio-to-score alignment. Although most score following methods assume that players have a precise score, most scores for guitarists have only melody and chord sequences without any beat patterns. An audio-visual beat tracking for guitarists is reported that improves the accuracy of beat detection. However, the result of this method is still insufficient because it uses only onset information, not pitch information, and because the hand tracking shows low accuracy. In this paper, we report a multimodal score following for a guitar performance, an extension of an audio-visual beat tracking method. The main difference is to use chord sequences to improve tracking of audio signals and depth information to improve tracking of guitar playing. Chord sequences are used for the calculation of chord correlation between the input and a score. Depth information is used in the guitar plane masking by three dimensional Hough transform, for the stable detection of a hand. Finally, the system extracts score positions and tempos by a particle-filter based integration of audio and visual features, The resulting score following system improves the tempo and the score position of a performance by 0.2 [sec] compared to an existing system. © 2012 IEEE.

    DOI

  • 音楽共演ロボット: 開始・終了キューの画像認識による人間のフルート奏者との実時間同期

    リムアンジェリカ, 水本武志, 大塚琢馬, 古谷ルイ賢造, 尾形哲也, 奥乃博

    情報処理学会論文誌   52 ( 12 ) 3599 - 3610  2011年12月

  • 音声対話システムにおける簡略表現認識のための自動語彙拡張

    森信介, 駒谷和範, 勝丸真樹, 尾形哲也, 奥乃博

    情報処理学会論文誌   52 ( 12 ) 3398 - 3407  2011年12月

  • 発語行為レベルの情報をユーザ発話の解釈に用いる音声対話システム

    駒谷和範, 松山匡子, 武田龍, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会論文誌   52 ( 12 ) 3374 - 3385  2011年12月

  • フレーズ置換のための調波非調波GMM・ NMF・残響推定に基づく音源分離・演奏合成

    安良岡直希, 吉岡拓也, 糸山克寿, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   52 ( 12 ) 3839 - 3852  2011年12月

  • An interactive musical ensemble with the NAO Thereminist

    Angelica Lim, 水本武志, 大塚琢馬, 糸原達彦, 中臺一博, 尾形哲也, 奥乃博

    第34回 AI チャレンジ研究会,人工知能学会    2011年12月

  • マルチロボットによるKinectを用いた同期合奏

    糸原達彦, 水本武志, Angelica Lim, 大塚琢馬, 中村圭佑, 長谷川雄二, 中臺一博, 尾形哲也, 奥乃博

    第34回 AI チャレンジ研究会, 人工知能学会   SIG-Challenge-B102-10, pp.4-49~4-54  2011年12月

  • ブラインド音源分離のためのInfinite Sparse Factor Analysisの複素拡張

    柳楽浩平, 高橋徹, 尾形哲也, 奥乃博

    第34回 AI チャレンジ研究会, 人工知能学会   SIG-Challenge-B102-9, pp.4-43~4-48  2011年12月

  • 音源定位手法MUSICのベイズ拡張

    大塚琢馬, 中臺一博, 尾形哲也, 奥乃博

    第34回 AI チャレンジ研究会, 人工知能学会   SIG-Challenge-B102-6, pp.4-25~4-30  2011年12月

  • Infinite Sparse Factor Analysis の複素拡張による音声信号のブラインド音源分離

    柳楽浩平, 高橋徹, 尾形哲也, 奥乃博

    日本音響学会関西支部第14回若手研究者交流研究発表会    2011年12月

  • More cowbell! A musical ensemble with the NAO thereminist

    Angelica Lim, Takeshi MIZUMOTO, Takuma OTSUKA, Tatsuhiko ITOHARA, Kazuhiro NAKADAI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-2011)    2011年09月  [査読有り]

  • Improved Statistical Model-Based Voice Activity Detection with Noise Reduction for the SIG-2 Humanoid Robot

    Uihyun Kim, Toru TAKAHASHI, Tetsuya OGATA, Hiroshi G. OKUNO

    日本ロボット学会第29回学術講演会     1Q1 - 7  2011年09月

  • Fast Incremental Probabilistic Surface Recognition for Robot Scene Understanding

    Louis Kenzo Cahier, Tetsuya OGATA, Hiroshi G. OKUNO

    日本ロボット学会第29回学術講演会     1Q2 - 3  2011年09月

  • 神経力学モデルによる文字列からの言語構造の自己組織化とロボット運動感 覚との統合

    尾形哲也, 日下航, 奥乃博

    日本ロボット学会第29回学術講演会     1A3 - 6  2011年09月

  • mproving social telepresence by converting emotional voice to robot gesture

    Angelica Lim, Tetsuya OGATA, Hiroshi G. OKUNO

    Angelica Lim, Tetsuya OGATA, Hiroshi G. OKUNO     1Q3 - 7  2011年09月

  • MUSIC 法を用いた音源定位のベイズ拡張

    大塚琢馬, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3A3 - 2  2011年09月

  • 対話データの再帰結合神経回路による 学習と相槌タイミング予測

    佐野正太郎, 西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第29回学術講演会     3O2 - 1  2011年09月

  • 神経力学モデルによる身体図式に基づく空間地図の獲得

    信田春満, 西出俊, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3N1 - 5  2011年09月

  • Improving social telepresence by converting emotional voice to robot gesture

    Angelica Lim, Tetsuya OGATA, Hiroshi G. OKUNO

    日本ロボット学会第29回学術講演会     1Q3 - 7  2011年09月

  • MUSIC 法を用いた音源定位のベイズ拡張

    大塚琢馬, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3A3 - 2  2011年09月

  • 対話データの再帰結合神経回路による 学習と相槌タイミング予測

    佐野正太郎, 西出俊, 奥乃博, 尾形哲也

    日本ロボット学会第29回学術講演会     3O2 - 1  2011年09月

  • 神経力学モデルによる身体図式に基づく空間地図の獲得

    信田春満, 西出俊, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3N1 - 5  2011年09月

  • 分散的ランドマーク環境における適応リズムによる移動ロボットの誤差低減

    小鷹研理, 尾形哲也, 菅野重樹

    日本ロボット学会第29回学術講演会     3N1 - 2  2011年09月

  • ノンパラメトリックベイズによる時間周波数領域における音声信号のブラインド音源分離

    柳楽浩平, 高橋徹, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3A2 - 5  2011年09月

  • 調波・非調波音源モデルを用いたマイク数以上の音源分離

    平澤恭治, 安良岡直希, 高橋徹, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3A2 - 4  2011年09月

  • パーティクルフィルタを用いたギター演奏の視聴覚統合ビートトラッキング

    糸原達彦, 大塚琢馬, 水本武志, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3A2 - 2  2011年09月

  • テルミン演奏ロボットのためのUnscented Kalman Filter による適応的音高制御

    水本武志, 尾形哲也, 奥乃博

    日本ロボット学会第29回学術講演会     3A2 - 1  2011年09月

  • ぺた語義:京大における Lisp を使ったプログラミング教育

    湯淺 太一, 奥乃 博, 尾形 哲也

    情報処理   52 ( 9 ) 1191 - 1194  2011年08月

    CiNii

  • 整合性基準に基づく多対多オーディオアライメント

    前澤陽, 糸山克寿, 尾形哲也, 奥乃博

    第91回音楽情報科学研究会, 情報処理学会     Vol.2011  2011年07月

  • 歌詞と音響特徴量を用いた楽曲印象軌跡推定法の設計と評価

    西川直毅, 糸山克寿, 藤原弘将, 後藤真孝, 尾形哲也, 奥乃博

    第91回音楽情報科学研究会,情報処理学会     Vol.2011  2011年07月

  • Evaluation of Spoken Dialogue System that uses Utterance Timing to Interprete User Utterances

    Kazunori KOMATANI, Kyoko MATSUYAMA, Ryu TAKEDA, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of International Workshop on Spoken Dialogue Systems (IWSDS2011)     315 - 325  2011年06月

  • 神経力学モデルによるロボットの言語・運動の統合的認知

    尾形哲也, 日下航, 奥乃博

    2011年度人工知能学会全国大会     3B1  2011年06月

  • 擬音語と環境音の音響的関係性を考慮した環境音to擬音語変換システム

    山川暢英, 北原鉄朗, 高橋徹, 尾形哲也, 奥乃博

    2011年度人工知能学会全国大会     1C2  2011年06月

  • Emergence of hierarchical structure mirroring linguistic composition in a recurrent neural network

    Wataru Hinoshita, Hiroaki Arie, Jun Tani, Hiroshi G. Okuno, Tetsuya Ogata

    Neural Networks   24 ( 4 ) 311 - 320  2011年05月  [査読有り]

     概要を見る

    We show that a Multiple Timescale Recurrent Neural Network (MTRNN) can acquire the capabilities to recognize, generate, and correct sentences by self-organizing in a way that mirrors the hierarchical structure of sentences: characters grouped into words, and words into sentences. The model can control which sentence to generate depending on its initial states (generation phase) and the initial states can be calculated from the target sentence (recognition phase). In an experiment, we trained our model over a set of unannotated sentences from an artificial language, represented as sequences of characters. Once trained, the model could recognize and generate grammatical sentences, even if they were not learned. Moreover, we found that our model could correct a few substitution errors in a sentence, and the correction performance was improved by adding the errors to the training sentences in each training iteration with a certain probability. An analysis of the neural activations in our model revealed that the MTRNN had self-organized, reflecting the hierarchical linguistic structure by taking advantage of the differences in timescale among its neurons: in particular, neurons that change the fastest represented "characters", those that change more slowly, "words", and those that change the slowest, "sentences". © 2011 Elsevier Ltd.

    DOI PubMed

  • ベース音高と和音特徴の統合に基づく和音系列認識

    須見康平, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   52 ( 4 ) 1803 - 1812  2011年04月

  • ロボット聴覚のためのMatching Pursuitによる複数環境音の同定

    山川鴨英, 北原鉄朗, 高橋徹, 尾形哲也, 奥乃 博

    情報処理学会第73回全国大会     6P - 3  2011年03月

  • L1ノルム最小化による劣決定音源分離のための線形計画と二次錐計画の比較評価

    平澤恭治, 武田龍, 高橋徹, 尾形哲也, 奥乃 博

    情報処理学会第73回全国大会     6P - 2  2011年03月

  • 音源数同定とブラインド音源分離を同時に行うinfinite ICA

    柳楽浩平, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     6P - 1  2011年03月

  • Audio-visual musical instrument recognition

    Angelica Lim, 中村圭佑, 中臺一博, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     5R - 9  2011年03月

  • 累積頻度重みを適用したパーティクルフィルタによる実時間楽譜追従

    大塚琢馬, 中臺一博, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     5R - 7  2011年03月

  • 伝達関数のスパース性仮定に基づく音楽音響信号中のディレイエフェクトブラインド推定

    阪上大地, 安良岡直希, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     5R - 5  2011年03月

  • 潜在的調波配分法に基づく隠れセミマルコフモデルを用いたベイズ的スコアアライメント

    前澤陽, 後藤真孝, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     5R - 4  2011年03月

  • 歌詞と音響特徴量を用いた楽曲の印象軌跡推定

    西川直毅, 糸山克寿, 藤原弘将, 後藤真孝, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     5R - 3  2011年03月

  • 調波パラメトリックNMFによる楽器演奏音響信号の分析合成

    安良岡直希, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     5R - 1  2011年03月

  • Classification of Harmonic and Textural Keyboard Playing Style Using Acoustic Features

    Jooyoung Ahn, 前澤陽, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     4C - 2,  2011年03月

  • Speaker Localization Using Two-Channel Microphone on the SIG-2 Humanoid Robot

    Uihyun Kim, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    情報処理学会第73回全国大会     4C - 1  2011年03月

  • 神経力学モデルの引込みによる相槌タイミング予測

    佐野正太郎, 尾形哲也, 日下航, 高橋徹, 奥乃博

    情報処理学会第73回全国大会     4P - 7  2011年03月

  • 誤認識頻発状況下で選択肢列挙を行う音声対話システムとその評価

    松山匡子, 駒谷和範, 武田龍, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     4P - 4  2011年03月

  • 神経回路モデルによる言語とロボット動作の相互連想学習

    日下航, 尾形哲也, 高橋徹, 奥乃博

    情報処理学会第73回全国大会     3Q - 1  2011年03月

  • 視聴覚統合ビートトラッキングを用いた音楽ロボットとギターとの合奏システム

    糸原達彦, 大塚琢馬, 水本武志, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     1ZB - 2  2011年03月

  • F0・音韻長・パワー制御による歌声らしさ・話声らしさの変化の評価

    阿曽慎平, 齋藤毅, 後藤真孝, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     2R - 6  2011年03月

  • Time-of-flight camera based Probabilistic Polygonal Mesh mapping

    Louis-Kenzo Cahier, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第73回全国大会     1T - 5  2011年03月

  • 再帰結合神経回路モデルへのスパース構造導入による学習能力の向上

    粟野皓光, 尾形哲也, 有江浩明, 谷淳, 高橋徹, 奥乃博

    情報処理学会第73回全国大会     1Q - 4  2011年03月

  • 神経力学モデルによる予測可能性を用いた身体識別

    信田春満, 尾形哲也, 日下航, 高橋徹, 奥乃博

    情報処理学会第73回全国大会     1Q - 1  2011年03月

  • Preface

    Tetsuya Ogata, Tetsuo Sawaragi, Tadahiro Taniguchi

    Advanced Robotics   25 ( 17 ) 2125 - 2126  2011年

    DOI

  • Robot Audition based on Multiple-Input Independent Component Analysis for Recognizing Barge-In Speech under Reverberation

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    JRSJ   27 ( 7 ) 782 - 792  2011年  [査読有り]

     概要を見る

    This paper presents a new method based on independent component analysis (ICA) for enhancing a target source and suppressing other interfering sound sources, supposed that the latter are known. The method can provides in a reverberant environment a barge-in-able robot audition system; that is, the user can talk to the robot at any time even when the robot speaks. Our method separates and dereverberates the user&#039;s speech and the robot&#039;s one by using Multiple Input ICA. The critical issue for real-time processing is to reduce the computational complexity of Multiple Input ICA to the linear order of the reverberation time, which has not been proposed so far. We attain it by exploit the property of the independence relationship between late observed signals and late speech signals. Experimental results show that 1) the computational complexity of our method is less than the na&amp;iuml;ve Multiple Input ICA method, and that 2) our method improves word correctness of automatic speech recognition under barge-in and reverberant situations; by at most 40 points for reverberation time of 240[ms] and 30 points for 670[ms].

    DOI CiNii

  • Phoneme Acquisition based on Vowel Imitation Model using Recurrent Neural Network and Physical Vocal Tract Model

    Hisashi Kanda, Tetsuya Ogata, Toru Takahashi, Kazunori Komatani, Hiroshi G. Okuno

    JRSJ   27 ( 7 ) 802 - 813  2011年  [査読有り]

     概要を見る

    This paper proposes a continuous vowel imitation system that explains the process of phoneme acquisition by infants from the dynamical systems perspective. Almost existing models concerning this process dealt with discrete phoneme sequences. Human infants, however, have no knowledge of phoneme innately. They perceive speech sounds as continuous acoustic signals. The imitation target of this study is continuous acoustic signals including unknown numbers and kinds of phonemes. The key ideas of the model are (1) the use of a physical vocal tract model called the Maeda model for embodying the motor theory of speech perception, (2) the use of a dynamical system called the Recurrent Neural Network with Parametric Bias (RNNPB) trained with both dynamics of the acoustic signals and articulatory movements of the Maeda model, and (3) the segmenting method of a temporal sequence using the prediction error of the RNNPB model. The experiments of our model demonstrated following results: (a) the self-organization of the vowel structure into attractors of RNNPB model, (b) the improvement of vowel imitation using movement of the Maeda model, and (c) the generation of clear vowels based on the bubbling process trained with a few random utterances. These results suggest that our model reflects the process of phoneme acquisition.

    DOI CiNii

  • Real-time audio-to-score alignment using particle filter for coplayer music robots

    Takuma Otsuka, Kazuhiro Nakadai, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Eurasip Journal on Advances in Signal Processing   2011  2011年  [査読有り]

     概要を見る

    Our goal is to develop a coplayer music robot capable of presenting a musical expression together with humans. Although many instrument-performing robots exist, they may have difficulty playing with human performers due to the lack of the synchronization function. The robot has to follow differences in humans' performance such as temporal fluctuations to play with human performers. We classify synchronization and musical expression into two levels: (1) melody level and (2) rhythm level to cope with erroneous synchronizations. The idea is as follows: When the synchronization with the melody is reliable, respond to the pitch the robot hears, when the synchronization is uncertain, try to follow the rhythm of the music. Our method estimates the score position for the melody level and the tempo for the rhythm level. The reliability of the score position estimation is extracted from the probability distribution of the score position. The experimental results demonstrate that our method outperforms the existing score following system in 16 songs out of 20 polyphonic songs. The error in the prediction of the score position is reduced by 69 on average. The results also revealed that the switching mechanism alleviates the error in the estimation of the score position. Copyright © 2011 Takuma Otsuka, et al.

    DOI

  • People Detection Based on Spatial Mapping of Friendliness and Floor Boundary Points for a Mobile Navigation Robot

    Tsuyoshi Tasaki, Fumio Ozaki, Nobuto Matsuhira, Tetsuya Ogata, Hiroshi G. Okuno

    Journal of Robotics   2011   1 - 10  2011年  [査読有り]

     概要を見る

    Navigation robots must single out partners requiring navigation and move in the cluttered environment where people walk around. Developing such robots requires two different people detections: detecting partners and detecting all moving people around the robots. For detecting partners, we design divided spaces based on the spatial relationships and sensing ranges. Mapping the friendliness of each divided space based on the stimulus from the multiple sensors to detect people calling robots positively, robots detect partners on the highest friendliness space. For detecting moving people, we regard objects’ floor boundary points in an omnidirectional image as obstacles. We classify obstacles as moving people by comparing movement of each point with robot movement using odometry data, dynamically changing thresholds to detect. Our robot detected 95.0% of partners while it stands by and interacts with people and detected 85.0% of moving people while robot moves, which was four times higher than previous methods did.

    DOI

  • Towards Written Text Recognition Based on Handwriting Experiences Using a Recurrent Neural Network.

    Shun Nishide, Jun Tani, Hiroshi G. Okuno, Tetsuya Ogata

    Adv. Robotics   25 ( 17 ) 2173 - 2187  2011年  [査読有り]

     概要を見る

    In this paper, we propose a model for recognizing written text through prediction of a handwriting sequence. The approach is based on findings in the brain sciences field. When recognizing written text, humans are said to unintentionally trace its handwriting sequence in their brains. Likewise, we aim to create a model that predicts a handwriting sequence from a static image of written text. The predicted handwriting sequence would be used to recognize the text. As the first step towards the goal, we created a model using neural networks, and evaluated the learning and recognition capability of the model using single Japanese characters. First, the handwriting image sequences for training are self-organized into image features using a self-organizing map. The self-organized image features are used to train the neuro-dynamics learning model. For recognition, we used both trained and untrained image sequences to evaluate the capability of the model to adapt to unknown data. The results of two experiments using 10 Japanese characters show the effectivity of the model. (C) Koninklijke Brill NV, Leiden and The Robotics Society of Japan, 2011

    DOI

  • Polyphonic audio-to-score alignment based on Bayesian latent harmonic allocation hidden Markov model

    Akira Maezawa, Hiroshi G. Okuno, Tetsuya Ogata, Masataka Goto

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings     185 - 188  2011年  [査読有り]

     概要を見る

    This paper presents a Bayesian method for temporally aligning a music score and an audio rendition. A critical problem in audio-to-score alignment is in dealing with the wide variety of timbre and volume of the audio rendition. In contrast with existing works that achieve this through ad-hoc feature design or careful training of tone models, we propose a Bayesian audio-to-score alignment method by modeling music performance as a Bayesian Hidden Markov Model, each state of which emits a Bayesian signal model based on Latent Harmonic Allocation. After attenuating reverberation, variational Bayes method is used to iteratively adapt the alignment, instrument tone model and the volume balance at each position of the score. The method is evaluated using sixty works of classical music of a variety of instrumentation ranging from solo piano to full orchestra. We verify that our method improves the alignment accuracy compared to dynamic time warping based on chroma vector for orchestral music, or our method employed in a maximum likelihood setting. © 2011 IEEE.

    DOI

  • Simultaneous processing of sound source separation and musical instrument identification using Bayesian spectral modeling

    Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings     3816 - 3819  2011年  [査読有り]

     概要を見る

    This paper presents a method of both separating audio mixtures into sound sources and identifying the musical instruments of the sources. A statistical tone model of the power spectrogram, called an integrated model, is defined and source separation and instrument identification are carried out on the basis of Bayesian inference. Since, the parameter distributions of the integrated model depend on each instrument, the instrument name is identified by selecting the one that has the maximum relative instrument weight. Experimental results showed correct instrument identification enables precise source separation even when many overtones overlap. © 2011 IEEE.

    DOI

  • Cluster self-organization of known and unknown environmental sounds using recurrent neural network

    Yang Zhang, Shun Nishide, Toru Takahashi, Hiroshi G. Okuno, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6791 LNCS ( PART 1 ) 167 - 175  2011年  [査読有り]

     概要を見る

    Our goal is to develop a system that is able to learn and classify environmental sounds for robots working in the real world. In the real world, two main restrictions pertain in learning. First, the system has to learn using only a small amount of data in a limited time because of hardware restrictions. Second, it has to adapt to unknown data since it is virtually impossible to collect samples of all environmental sounds. We used a neuro-dynamical model to build a prediction and classification system which can self-organize sound classes into parameters by learning samples. The proposed system searches space of parameters for classifying. In the experiment, we evaluated the accuracy of classification for known and unknown sound classes. © 2011 Springer-Verlag.

    DOI

  • Robot with two ears listens to more than two simultaneous utterances by exploiting harmonic structures

    Yasuharu Hirasawa, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6703 LNAI ( PART 1 ) 348 - 358  2011年

     概要を見る

    In real-world situations, people often hear more than two simultaneous sounds. For robots, when the number of sound sources exceeds that of sensors, the situation is called under-determined, and robots with two ears need to deal with this situation. Some studies on under-determined sound source separation use L1-norm minimization methods, but the performance of automatic speech recognition with separated speech signals is poor due to its spectral distortion. In this paper, a two-stage separation method to improve separation quality with low computational cost is presented. The first stage uses a L1-norm minimization method in order to extract the harmonic structures. The second stage exploits reliable harmonic structures to maintain acoustic features. Experiments that simulate three utterances recorded by two microphones in an anechoic chamber show that our method improves speech recognition correctness by about three points and is fast enough for real-time separation. © 2011 Springer-Verlag.

    DOI

  • Environmental sound recognition for robot audition using matching-pursuit

    Nobuhide Yamakawa, Toru Takahashi, Tetsuro Kitahara, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6704 LNAI ( PART 2 ) 1 - 10  2011年

     概要を見る

    Our goal is to achieve a robot audition system that is capable of recognizing multiple environmental sounds and making use of them in human-robot interaction. The main problems in environmental sound recognition in robot audition are: (1) recognition under a large amount of background noise including the noise from the robot itself, and (2) the necessity of robust feature extraction against spectrum distortion due to separation of multiple sound sources. This paper presents the environmental recognition of two sound sources fired simultaneously using matching pursuit (MP) with the Gabor wavelet, which extracts salient audio features from a signal. The two environmental sounds come from different directions, and they are localized by multiple signal classification and, using their geometric information, separated by geometric source separation with the aid of measured head-related transfer functions. The experimental results show the noise-robustness of MP although the performance depends on the properties of the sound sources. © 2011 Springer-Verlag.

    DOI

  • Fast and simple iterative algorithm of Lp-norm minimization for under-determined speech separation

    Yasuharu Hirasawa, Naoki Yasuraoka, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH     1745 - 1748  2011年  [査読有り]

     概要を見る

    This paper presents an efficient algorithm to solve Lp-norm minimization problem for under-determined speech separation; that is, for the case that there are more sound sources than microphones. We employ an auxiliary function method in order to derive update rules under the assumption that the amplitude of each sound source follows generalized Gaussian distribution. Experiments reveal that our method solves the L1-norm minimization problem ten times faster than a general solver, and also solves Lp-norm minimization problem efficiently, especially when the parameter p is small; when p is not more than 0.7, it runs in real-time without loss of separation quality. Copyright © 2011 ISCA.

  • Bayesian extension of MUSIC for sound source localization and tracking

    Takuma Otsuka, Kazuhiro Nakadai, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH     3109 - 3112  2011年  [査読有り]

     概要を見る

    This paper presents a Bayesian extension of MUSIC-based sound source localization (SSL) and tracking method. SSL is important for distant speech enhancement and simultaneous speech separation for improving speech recognition, as well as for auditory scene analysis by mobile robots. One of the draw- backs of existing SSL methods is the necessity of careful param- eter tunings, e.g., the sound source detection threshold depend- ing on the reverberation time and the number of sources. Our contribution consists of (1) automatic parameter estimation in the variational Bayesian framework and (2) tracking of sound sources with reliability. Experimental results demonstrate our method robustly tracks multiple sound sources in a reverberant environment with RT20 = 840 (ms). Copyright © 2011 ISCA.

  • Particle-filter based audio-visual beat-tracking for music robot ensemble with human guitarist

    Tatsuhiko Itohara, Takuma Otsuka, Takeshi Mizumoto, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     118 - 124  2011年

     概要を見る

    This paper presents an audio-visual beat-tracking method for ensemble robots with a human guitarist. Beat-tracking, or estimation of tempo and beat times of music, is critical to the high quality of musical ensemble performance. Since a human plays the guitar in out-beat in back beat and syncopation, the main problems of beat-tracking of a human's guitar playing are twofold: tempo changes and varying note lengths. Most conventional methods have not addressed human's guitar playing. Therefore, they lack the adaptation of either of the problems. To solve the problems simultaneously, our method uses not only audio but visual features. We extract audio features with Spectro-Temporal Pattern Matching (STPM) and visual features with optical flow, mean shift and Hough transform. Our beat-tracking estimates tempo and beat time using a particle filter; both acoustic feature of guitar sounds and visual features of arm motions are represented as particles. The particle is determined based on prior distribution of audio and visual features, respectively Experimental results confirm that our integrated audio-visual approach is robust against tempo changes and varying note lengths. In addition, they also show that estimation convergence rate depends only a little on the number of particles. The real-time factor is 0.88 when the number of particles is 200, and this shows out method works in real-time. © 2011 IEEE.

    DOI

  • Improvement of speaker localization by considering multipath interference of sound wave for binaural robot audition

    Ui Hyun Kim, Takeshi Mizumoto, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     2910 - 2915  2011年  [査読有り]

     概要を見る

    This paper presents an improved speaker localization method based on the generalized cross-correlation (GCC) method weighted by the phase transform (PHAT) for binaural robot audition. The problem with the conventional direction-of-arrival (DOA) estimation based on the GCC-PHAT method is a multipath interference whereby a sound wave travels to microphones via the front-head path and the back-head path in binaural robot audition. This paper describes a new time delay factor for the GCC-PHAT method to compensate multipath interference on the assumption of spherical robot head. In addition, the restriction of the time difference of arrival (TDOA) estimation by the sampling frequency is also solved by applying the maximum likelihood (ML) estimation in frequency domain. Experiments conducted in the SIG-2 humanoid robot show that the proposed method reduces localization errors by 17.8 degrees on average and by over 35 degrees in side directions comparing to the conventional DOA estimation. © 2011 IEEE.

    DOI

  • Classification of known and unknown environmental sounds based on self-organized space using a recurrent neural network

    Yang Zhang, Tetsuya Ogata, Shun Nishide, Toru Takahashi, Hiroshi G. Okuno

    Advanced Robotics   25 ( 17 ) 2127 - 2141  2011年  [査読有り]

     概要を見る

    Our goal is to develop a system to learn and classify environmental sounds for robots working in the real world. In the real world, two main restrictions pertain in learning. (i) Robots have to learn using only a small amount of data in a limited time because of hardware restrictions. (ii) The system has to adapt to unknown data since it is virtually impossible to collect samples of all environmental sounds. We used a neuro-dynamical model to build a prediction and classification system. This neuro-dynamical model can self-organize sound classes into parameters by learning samples. The sound classification space, constructed by these parameters, is structured for the sound generation dynamics and obtains clusters not only for known classes, but also unknown classes. The proposed system searches on the basis of the sound classification space for classifying. In the experiment, we evaluated the accuracy of classification for both known and unknown sound classes. © 2011 Koninklijke Brill NV, Leiden.

    DOI

  • Handwriting prediction based character recognition using recurrent neural network

    Shun Nishide, Hiroshi G. Okuno, Tetsuya Ogata, Jun Tani

    Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics     2549 - 2554  2011年  [査読有り]

     概要を見る

    Humans are said to unintentionally trace handwriting sequences in their brains based on handwriting experiences when recognizing written text. In this paper, we propose a model for predicting handwriting sequence for written text recognition based on handwriting experiences. The model is first trained using image sequences acquired while writing text. The image features of sequences are self-organized from the images using Self-Organizing Map. The feature sequences are used to train a neuro-dynamics learning model. For recognition, the text image is input into the model for predicting the handwriting sequence and recognition of the text. We conducted two experiments using ten Japanese characters. The results of the experiments show the effectivity of the model. © 2011 IEEE.

    DOI

  • Incremental bayesian audio-to-score alignment with flexible harmonic structure models

    Takuma Otsuka, Kazuhiro Nakadai, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 12th International Society for Music Information Retrieval Conference, ISMIR 2011     525 - 530  2011年  [査読有り]

     概要を見る

    Music information retrieval, especially the audio-to-score alignment problem, often involves a matching problem between the audio and symbolic representations. We must cope with uncertainty in the audio signal generated from the score in a symbolic representation such as the variation in the timbre or temporal fluctuations. Existing audio-to-score alignment methods are sometimes vulnerable to the uncertainty in which multiple notes are simultaneously played with a variety of timbres because these methods rely on static observation models. For example, a chroma vector or a fixed harmonic structure template is used under the assumption that musical notes in a chord are all in the same volume and timbre. This paper presents a particle filterbased audio-to-score alignment method with a flexible observation model based on latent harmonic allocation. Our method adapts to the harmonic structure for the audio-toscore matching based on the observation of the audio signal through Bayesian inference. Experimental results with 20 polyphonic songs reveal that our method is effective when more number of instruments are involved in the ensemble. © 2011 International Society for Music Information Retrieval.

  • Converting emotional voice to motion for robot telepresence

    Angelica Lim, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE-RAS International Conference on Humanoid Robots     472 - 479  2011年  [査読有り]

     概要を見る

    In this paper we present a new method for producing affective motion for humanoid robots. The NAO robot, like other humanoids, does not possess facial features to convey emotion. Instead, our proposed system generates pose-independent robot movement using a description of emotion through speed, intensity, regularity and extent (DESIRE). We show how the DESIRE framework can link the emotional content of voice and gesture, without the need for an emotion recognition system. Our results show that DESIRE movement can be used to effectively convey at least four emotions with user agreement 60-75%, and that voices converted to motion through SIRE maintained the same emotion significantly higher than chance, even across cultures (German to Japanese). Additionally, portrayals recognized as happiness were rated significantly easier to understand with motion over voice alone. © 2011 IEEE.

    DOI

  • A musical mood trajectory estimation method using lyrics and acoustic features

    Naoki Nishikawa, Katsutoshi Itoyama, Hiromasa Fujihara, Masataka Goto, Tetsuya Ogata, Hiroshi G. Okuno

    MM'11 - Proceedings of the 2011 ACM Multimedia Conference and Co-Located Workshops - MIRUM 2011 Workshop, MIRUM'11     51 - 56  2011年  [査読有り]

     概要を見る

    In this paper, we present a new method that represents an overall musical time-varying impression of a song by a pair of mood trajectories estimated from lyrics and audio signals. The mood trajectory of the lyrics is obtained by using the probabilistic latent semantic analysis (PLSA) to estimate topics (representing impressions) from words in the lyrics. The mood trajectory of the audio signals is estimated from acoustic features by using the multiple linear regression analysis. In our experiments, the mood trajectories of 100 songs in Last.fm's Best of 2010 were estimated. The detailed analysis of the 100 songs confirms that acoustic features provide more accurate mood trajectory and the 21% resulting mood trajectories are matched to realistic musical mood available at Last.fm. © 2011 ACM.

    DOI

  • Use of a sparse structure to improve learning performance of recurrent neural networks

    Hiromitsu Awano, Shun Nishide, Hiroaki Arie, Jun Tani, Toru Takahashi, Hiroshi G. Okuno, Tetsuya Ogata

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   7064 LNCS ( PART 3 ) 323 - 331  2011年  [査読有り]

     概要を見る

    The objective of our study is to find out how a sparse structure affects the performance of a recurrent neural network (RNN). Only a few existing studies have dealt with the sparse structure of RNN with learning like Back Propagation Through Time (BPTT). In this paper, we propose a RNN with sparse connection and BPTT called Multiple time scale RNN (MTRNN). Then, we investigated how sparse connection affects generalization performance and noise robustness. In the experiments using data composed of alphabetic sequences, the MTRNN showed the best generalization performance when the connection rate was 40%. We also measured sparseness of neural activity and found out that sparseness of neural activity corresponds to generalization performance. These results means that sparse connection improved learning performance and sparseness of neural activity would be used as metrics of generalization performance. © 2011 Springer-Verlag.

    DOI

  • Exploring movable space using rhythmical active touch in disordered obstacle environment

    Kenri Kodaka, Tetsuya Ogata, Hirotaka Ohta, Shigeki Sugano

    2011 IEEE/SICE International Symposium on System Integration, SII 2011     485 - 490  2011年

     概要を見る

    We propose a novel navigation system for adaptively exploring an obstacle space using diverse ways of touching an object. Conventional navigation models are typically based on the avoidance of obstacles, i.e., avoiding collision. However, actual disordered space may be full of various kinds of obstacles. To reach a destination in such a space, a robot requires an active approach for avoiding a deadlock with obstacles or changing the obstacle configuration to find an open space using diverse ways of touching an object. We solved this problem by generating locally diverse moving patterns by using an action model with rhythmical oscillation in addition to a localization model using a particle filter. The proposed model was demonstrated to be effective through an experiment where a robot navigated to a destination behind partially movable obstacles using rhythmical active touch. © 2011 IEEE.

    DOI

  • Predicting listener back-channels for human-agent interaction using neuro-dynamical model

    Shotaro Sano, Shun Nishide, Hiroshi G. Okuno, Tetsuya Ogata

    2011 IEEE/SICE International Symposium on System Integration, SII 2011     18 - 23  2011年

     概要を見る

    The goal of our work is to create natural verbal interaction between humans and speech dialogue agents. In this paper, we focus on generations of back-channel for speech dialogue agents the same way humans do. To create such a system, the system needs to predict the appropriate timing of back-channel on the basis of the human's speech. For the prediction model, we use a neuro-dynamical system called a multiple timescale recurrent neural network (MTRNN). The model is trained using an actual corpus of a poster session of the IMADE project using the presenter's prosodic and visual information as features. Using the model, we conducted back-channel timing prediction experiments. The results showed that our system could predict back-channel timing about 0.5 seconds before generation of back-channel response. Comparing the results with the actual back-channel timing in the corpus, the system showed 37.1% of recall, 31.7% of precision, and 34.2% of F-measure. These results show the model to effectively predict and generate back-channel responses. © 2011 IEEE.

    DOI

  • Identification of self-body based on dynamic predictability using neuro-dynamical system

    Harumitsu Nobuta, Shun Nishide, Hiroshi G. Okuno, Tetsuya Ogata

    2011 IEEE/SICE International Symposium on System Integration, SII 2011     256 - 261  2011年

     概要を見る

    The goal of our work is to acquire an internal model through a robot's experience. The internal model has the ability for mutual conversion between motor commands and movement of the body (e.g. hand) in view. Unlike other works, which assume the robot's body to be extracted in its view, we assume that external moving objects are also included in its view. We introduce predictability as a measure to segregate such objects from the robot's body: the robot's body is predictable while moving objects are not. Prediction is conducted using a neuro-dynamical system called the multiple timescales recurrent neural network (MTRNN). The prediction results of the robot's body are compared with the actual motion to distinguish the robot's body from other objects. For evaluation, we conducted an experiment with the robot moving its hand while moving objects were in view. The results of the experiment showed that the prediction of the robot's hand is 3.86 times as accurate as that of others on average. These results show the effectiveness of using predictability as a measure to acquire an internal model in an environment that includes both a robot's body and other moving objects in view. © 2011 IEEE.

    DOI

  • Selecting help messages by using robust grammar verification for handling out-of-grammar utterances in spoken dialogue systems

    Kazunori Komatani, Yuichiro Fukubayashi, Satoshi Ikeda, Tetsuya Ogata, Hiroshi G. Okuno

    IEICE Transactions on Information and Systems   E93-D ( 12 ) 3359 - 3367  2010年12月  [査読有り]

     概要を見る

    We address the issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages. Help message generation for OOG utterances is a challenge because language understanding based on automatic speech recognition (ASR) of OOG utterances is usually erroneous; important words are often misrecognized or missing from such utterances. Our grammar verification method uses a weighted finite-state transducer, to accurately identify the grammar rule that the user intended to use for the utterance, even if important words are missing from the ASR results. We then use a ranking algorithm, RankBoost, to rank help message candidates in order of likely usefulness. Its features include the grammar verification results and the utterance history representing the user's experience. Copyright © 2010.

    DOI

  • コミュニケーションロボットWAMOEBA-3の開発

    菅野重樹, 尾形哲也, 菅佑樹, 西佑起

    第11回システムインテグレーション部門講演会 (SI2010), 計測自動制御学会     1E3 - 3  2010年12月

  • 自己形態主張型カスタマイズロボットの開発

    山崎由美子, 守良真, 菅佑樹, 尾形哲也, 菅野重樹

    第11回システムインテグレーション部門講演会 (SI2010),計測自動制御学会     1E3 - 4  2010年12月

  • 再帰神経回路による環境音の構造化と識別

    張陽, 尾形哲也, 高橋徹, 奥乃博

    第11回システムインテグレーション部門講演会 (SI2010), 計測自動制御学会     2I1 - 5  2010年12月

  • テルミンの音高・音量特性のモデルに基づくテルミン演奏ロボットの開発

    水本武志, 辻野広司, 高橋徹, 駒谷和範, 尾形哲也

    情報処理学会論文誌   51 ( 10 ) 2008 - 2019  2010年10月

    CiNii

  • Inter-modality mapping in robot with recurrent neural network

    Tetsuya Ogata, Shun Nishide, Hideki Kozima, Kazunori Komatani, Hiroshi G. Okuno

    Pattern Recognition Letters   31 ( 12 ) 1560 - 1569  2010年09月  [査読有り]

     概要を見る

    A system for mapping between different sensory modalities was developed for a robot system to enable it to generate motions expressing auditory signals and sounds generated by object movement. A recurrent neural network model with parametric bias, which has good generalization ability, is used as a learning model. Since the correspondences between auditory signals and visual signals are too numerous to memorize, the ability to generalize is indispensable. This system was implemented in the "Keepon" robot, and the robot was shown horizontal reciprocating or rotating motions with the sound of friction and falling or overturning motion with the sound of collision by manipulating a box object. Keepon behaved appropriately not only from learned events but also from unknown events and generated various sounds in accordance with observed motions. © 2009 Elsevier B.V. All rights reserved.

    DOI

  • 多重奏音響信号中の歌唱音声の歌詞を自由に差し替える歌詞置換システム

    安良岡直希, 糸山克寿, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    音響学会秋季学術講演会    2010年09月

  • Multimodal gesture recognition for robot musical accompaniment

    Lim Angelica, Mizumoto Takeshi, Cahier Louis-Kenzo, Otsuka Takuma, Takahashi Toru, Ogata Tetsuya, Okuno Hiroshi G

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1C1 - 4  2010年09月

  • 調波構造を用いたL1ノルム最小化に基づく劣決定音源分離手法の性能評価

    平澤恭治, 高橋徹, 尾形哲也, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1H2 - 3  2010年09月

  • ロボット聴覚のためのMatching-Pursuitによる環境音の分離音認識

    山川暢英, 高橋徹, 北原鉄朗, 尾形哲也, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1H2 - 4  2010年09月

  • Dynamic Recognition of Enviromental Sounds with Recurrent Neural Network

    張陽, 尾形哲也, 高橋徹, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1H2 - 7  2010年09月

  • リサンプル-ブロック処理と並列化に基づくICAの実時間実装

    武田龍, 中臺一博, 高橋徹, 尾形哲也, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1H3 - 1,  2010年09月

  • 打楽器とロボットの合奏のための結合振動子モデルに基づく打撃時刻予測手法

    水本武志, 中臺一博, 大塚琢馬, 高橋徹, 尾形哲也, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1H3 - 2  2010年09月

  • 音楽ロボットのためのパーティクルフィルタを用いたテンポ・楽譜追従手法

    大塚琢馬, 中臺一博, 高橋徹, 尾形哲也, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     1H3 - 6  2010年09月

  • Probabilistic polygonal mesh for 3D SLAM

    Cahier Louis-Kenzo, Takahashi Toru, Ogata Tetsuya, Okuno Hiroshi G

    日本ロボット学会第28回学術講演会, 名古屋工業大学     2B2 - 4  2010年09月

  • 確信度を用いた物体配置作業における人間ロボット協調

    粟野皓光, 尾形哲也, 西出俊, 高橋徹, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     3J1 - 6  2010年09月

  • 自己形態主張型カスタマイズロボットの開発?音を用いたカスタマイズ要求主張行動の検証?

    山崎由美子, 守良真, 近藤裕樹, 菅祐樹, 尾形哲也, 菅野重樹

    日本ロボット学会第28回学術講演会, 名古屋工業大学     3C2 - 5  2010年09月

  • 能動知覚経験に基づく物体特徴量の自己組織化と予測信頼性に基づく動作生成

    西出俊, 尾形哲也, 谷淳, 高橋徹, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     3A2 - 5  2010年09月

  • MTRNNを用いた階層的言語構造の創発

    日下航, 有江浩明, 谷淳, 尾形哲也, 高橋徹, 奥乃博

    日本ロボット学会第28回学術講演会, 名古屋工業大学     3A2 - 7  2010年09月

  • バージイン許容音声対話システムにおけるユーザ発話の分析と指示対象同定への応用

    松山匡子, 駒谷和範, 武田龍, 尾形哲也, 奥乃 博

    第 回音声言語情報処理研究会 情処研報,情報処理学会   Vol.2010 ( No.10 ) 2010  2010年07月

  • 多重奏音響信号中の演奏をユーザー指定の旋律に差し替えるフレーズ置換システム

    安良岡直希, 糸山克寿, 吉岡拓也, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報科学研究会, つくば, 情報処理学会   Vol.2010-MUS-, No., pp.  2010年07月

  • SpeakBySinging: 歌声を話声に変換する話声合成システム

    阿曽慎平, 齋藤毅, 後藤真孝, 糸山克寿, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報科学研究会,つくば, 情報処理学会   Vol.2010-MUS-86, No., pp.  2010年07月

  • 複数の言語モデルと言語理解モデルによる音声理解の高精度化

    勝丸 真樹, 中野 幹生, 駒谷 和範, 船越 孝太郎, 辻野 広司, 尾形 哲也, 奥乃 博

    電子情報通信学会論文誌. D, 情報・システム = The IEICE transactions on information and systems (Japanese edition)   93 ( 6 ) 879 - 888  2010年06月

     概要を見る

    本論文では,音声対話システムにおいて,複数の言語モデルと複数の言語理解モデルを用いることで,高精度な音声理解を行う手法について述べる.ユーザの発話によって適した言語モデル・言語理解モデルの組合せは異なることから,単一の音声理解方式で様々な発話に対して高精度な音声理解を実現することは難しい.そこで本論文では,まず,複数の言語モデルと言語理解モデルを用いて複数の理解結果を得ることで,理解結果の候補を得る.次に,得られた複数の理解結果に対して,ロジスティック回帰に基づき発話単位の信頼度を付与し,その信頼度が最も高い理解結果を選択する.本論文では,言語モデルとして文法モデルとN-gramモデルの2種類,言語理解モデルとしてFinite-State Transducer(FST)とWeighted FST(WFST),Keyphrase-Extractorの3種類を用いた.評価実験では,言語モデル・言語理解モデルのいずれかを複数用いた場合と比較して,コンセプト理解精度の向上が得られた.また,従来のROVER法による理解結果の統合と比較し,本手法の有効性を示した.

    CiNii

  • RNNを備えた2体のロボット間における身体性に基づいた動的コミュニケーションの創発

    日下航, 尾形哲也, 小嶋秀樹, 高橋徹, 奥乃博

    日本ロボット学会誌   28 ( 4 ) 532 - 543  2010年05月

  • "Score Following by Particle Filtering for Music Robots, "

    大塚琢馬, 中臺一博, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     5R - 7  2010年03月

  • 調波非調波GMMに基づくMIDI演奏音響信号に対する音色・演奏表情操作

    安良岡直希, 糸山克寿, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.5T - 5  2010年03月

  • クラシック音楽理解能力拡張インターフェイスのための同音旋律音量推定手法と主旋律推定への応用

    前澤陽, 後藤真孝, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.3T - 1  2010年03月

  • 楽器音イコライザによる楽曲音響特徴変動と類似楽曲検索への応用

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.6J - 6  2010年03月

  • F0・振幅・音韻長の制御により歌声を話声に変換する話声合成システムSpeakBySinging

    阿曽慎平, 齋藤毅, 後藤真孝, 糸山克寿, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.6U - 1  2010年03月

  • MTRNNを用いた単語と文法の階層的自己組織化による文の認識・生成

    日下航, 有江浩明, 谷淳, 尾形哲也, 高橋徹, 駒谷和範, 奥乃博

    情報処理学会第72回全国大会     p.6W - 8  2010年03月

  • RNNを用いた行為予測による人間とロボットの協調物体配置

    粟野皓光, 尾形哲也, 高橋徹, 駒谷和範, 奥乃博

    情報処理学会第72回全国大会     p.5V - 6  2010年03月

  • 環境音から擬音語への自動変換における特徴量抽出法の検討

    山川暢英, 北原鉄朗, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.3U - 9  2010年03月

  • 実環境音声認識のためのロボット聴覚システム開発とパラメータチューニング

    高橋徹, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.6J - 8  2010年03月

  • ロボット音声対話におけるSemi-blind ICAを用いた自己発話キャンセル

    武田龍, 中臺一博, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.6J - 7  2010年03月

  • 複数自由度を用いて音高特性モデルに基づく音高制御を行うテルミン演奏ロボットの開発

    水本武志, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.6T - 8  2010年03月

  • スペクトル推定を用いたマイク数以上の同時発話に対する音声認識

    平澤恭治, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.3U - 7  2010年03月

  • ユーザの文法知識を状態に加えたPOMDPに基づく音声対話システム

    穐山空道, 駒谷和範, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.5U - 9  2010年03月

  • バージイン許容音声対話におけるLSMによる許容発話範囲の拡張

    松山匡子, 駒谷和範, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.5U - 9  2010年03月

  • Robot Musical Accompaniment: Real-time Synchronization using Visual Cue Recognition

    Angelica Lim, 水本武志, 大塚琢馬, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.6T - 7  2010年03月

  • 複数の言語モデルと言語理解モデルによる音声理解手法のラピッドプロトタイピングへの適用

    勝丸真樹, 駒谷和範, 中野幹生, 船越孝太郎, 辻野広司, 尾形哲也, 奥乃博

    情報処理学会第72回全国大会     p.3U - 2  2010年03月

  • Soft missing-feature mask generation for robot audition.

    Toru Takahashi 0001, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Paladyn J. Behav. Robotics   1 ( 1 ) 37 - 47  2010年

    DOI

  • Voice-awareness control for a humanoid robot consistent with its body posture and movements.

    Takuma Otsuka, Kazuhiro Nakadai, Toru Takahashi 0001, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Paladyn J. Behav. Robotics   1 ( 1 ) 80 - 88  2010年

    DOI

  • Upper-limit evaluation of robot audition based on ICA-BSS in multi-source, barge-in and highly reverberant conditions

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     4366 - 4371  2010年  [査読有り]

     概要を見る

    This paper presents the upper-limit evaluation of robot audition based on ICA-BSS in multi-source, barge-in and highly reverberant conditions. The goal is that the robot can automatically distinguish a target speech from its own speech and other sound sources in a reverberant environment. We focus on the multi-channel semi-blind ICA (MCSB-ICA), which is one of the sound source separation methods with a microphone array, to achieve such an audition system because it can separate sound source signals including reverberations with few assumptions on environments. The evaluation of MCSB-ICA has been limited to robot's speech separation and reverberation separation. In this paper, we evaluate MCSB-ICA extensively by applying it to multi-source separation problems under common reverberant environments. Experimental results prove that MCSB-ICA outperforms conventional ICA by 30 points in automatic speech recognition performance. ©2010 IEEE.

    DOI

  • Improvement in listening capability for humanoid robot HRP-2

    Toru Takahashi, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     470 - 475  2010年  [査読有り]

     概要を見る

    This paper describes improvement of sound source separation for a simultaneous automatic speech recognition (ASR) system of a humanoid robot. A recognition error in the system is caused by a separation error and interferences of other sources. In separability, an original geometric source separation (GSS) is improved. Our GSS uses a measured robot's head related transfer function (HRTF) to estimate a separation matrix. As an original GSS uses a simulated HRTF calculated based on a distance between microphone and sound source, there is a large mismatch between the simulated and the measured transfer functions. The mismatch causes a severe degradation of recognition performance. Faster convergence speed of separation matrix reduces separation error. Our approach gives a nearer initial separation matrix based on a measured transfer function from an optimal separation matrix than a simulated one. As a result, we expect that our GSS improves the convergence speed. Our GSS is also able to handle an adaptive step-size parameter. These new features are added into open source robot audition software (OSS) called "HARK" which is newly updated as version 1.0.0. The HARK has been installed on a HRP-2 humanoid with an 8-element microphone array. The listening capability of HRP-2 is evaluated by recognizing a target speech signal which is separated from a simultaneous speech signal by three talkers. The word correct rate (WCR) of ASR improves by 5 points under normal acoustic environments and by 10 points under noisy environments. Experimental results show that HARK 1.0.0 improves the robustness against noises. ©2010 IEEE.

    DOI

  • Improving identification accuracy by extending acceptable utterances in spoken dialogue system using barge-in timing

    Kyoko Matsuyama, Kazunori Komatani, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6097 LNAI ( PART 2 ) 585 - 594  2010年  [査読有り]

     概要を見る

    We describe a novel dialogue strategy enabling robust interaction under noisy environments where automatic speech recognition (ASR) results are not necessarily reliable. We have developed a method that exploits utterance timing together with ASR results to interpret user intention, that is, to identify one item that a user wants to indicate from system enumeration. The timing of utterances containing referential expressions is approximated by Gamma distribution, which is integrated with ASR results by expressing both of them as probabilities. In this paper, we improve the identification accuracy by extending the method. First, we enable interpretation of utterances including ordinal numbers, which appear several times in our data collected from users. Then we use proper acoustic models and parameters, improving the identification accuracy by 4.0% in total. We also show that Latent Semantic Mapping (LSM) enables more expressions to be handled in our framework. © 2010 Springer-Verlag.

    DOI

  • Violin fingering estimation based on violin pedagogical fingering model constrained by bowed sequence estimation from audio input

    Akira Maezawa, Katsutoshi Itoyama, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6098 LNAI ( PART 3 ) 249 - 259  2010年  [査読有り]

     概要を見る

    This work presents an automated violin fingering estimation method that facilitates a student violinist acquire the "sound" of his/her favorite recording artist created by the artist's unique fingering. Our method realizes this by analyzing an audio recording played by the artist, and recuperating the most playable fingering that recreates the aural characteristics of the recording. Recovering the aural characteristics requires the bowed string estimation of an audio recording, and using the estimated result for optimal fingering decision. The former requires high accuracy and robustness against the use of different violins or brand of strings; and the latter needs to create a natural fingering for the violinist. We solve the first problem by detecting estimation errors using rule-based algorithms, and by adapting the estimator to the recording based on mean normalization. We solve the second problem by incorporating, in addition to generic stringed-instrument model used in existing studies, a fingering model that is based on pedagogical practices of violin playing, defined on a sequence of two or three notes. The accuracy of the bowed string estimator improved by 21 points in a realistic situation (38% → 59%) by incorporating error correction and mean normalization. Subjective evaluation of the optimal fingering decision algorithm by seven violinists on 22 musical excerpts showed that compared to the model used in existing studies, our proposed model was preferred over existing one (p=0.01), but no significant preference towards proposed method defined on sequence of two notes versus three notes was observed (p=0.05). © 2010 Springer-Verlag.

    DOI

  • Music-ensemble robot that is capable of playing the Theremin while listening to the accompanied music

    Takuma Otsuka, Takeshi Mizumoto, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6096 LNAI ( PART 1 ) 102 - 112  2010年  [査読有り]

     概要を見る

    Our goal is to achieve a musical ensemble among a robot and human musicians where the robot listens to the music with its own microphones. The main issues are (1) robust beat-tracking since the robot hears its own generated sounds in addition to the accompanied music, and (2) robust synchronizing its performance with the accompanied music even if humans' musical performance fluctuates. This paper presents a music-ensemble Thereminist robot implemented on the humanoid HRP-2 with the following three functions: (1) self-generated Theremin sound suppression by semi-blind Independent Component Analysis, (2) beat tracking robust against tempo fluctuation in humans' performance, and (3) feedforward control of Theremin pitch. Experimental results with a human drummer show the capability of this robot for the adaptation to the temporal fluctuation in his performance. © 2010 Springer-Verlag.

    DOI

  • Recognition and generation of sentences through self-organizing linguistic hierarchy using MTRNN

    Wataru Hinoshita, Hiroaki Arie, Jun Tani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6098 LNAI ( PART 3 ) 42 - 51  2010年  [査読有り]

     概要を見る

    We show that a Multiple Timescale Recurrent Neural Network (MTRNN) can acquire the capabilities of recognizing and generating sentences by self-organizing a hierarchical linguistic structure. There have been many studies aimed at finding whether a neural system such as the brain can acquire languages without innate linguistic faculties. These studies have found that some kinds of recurrent neural networks could learn grammar. However, these models could not acquire the capability of deterministically generating various sentences, which is an essential part of language functions. In addition, the existing models require a word set in advance to learn the grammar. Learning languages without previous knowledge about words requires the capability of hierarchical composition such as characters to words and words to sentences, which is the essence of the rich expressiveness of languages. In our experiment, we trained our model to learn language using only a sentence set without any previous knowledge about words or grammar. Our experimental results demonstrated that the model could acquire the capabilities of recognizing and deterministically generating grammatical sentences even if they were not learned. The analysis of neural activations in our model revealed that the MTRNN had self-organized the linguistic structure hierarchically by taking advantage of differences in the time scale among its neurons, more concretely, neurons that change the fastest represented "characters," those that change more slowly represented "words," and those that change the slowest represented "sentences." © 2010 Springer-Verlag.

    DOI

  • Design and implementation of two-level synchronization for an interactive music robot

    Takuma Otsuka, Kazuhiro Nakadai, Tom Takahashi, Kazunori Komatanj, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the National Conference on Artificial Intelligence   2   1238 - 1244  2010年  [査読有り]

     概要を見る

    Our goal is to develop an interactive music robot, i.e., a robot that presents a musical expression together with humans. A music interaction requires two important functions: synchronization with the music and musical expression, such as singing and dancing. Many instrument-performing robots are only capable of the latter function, they may have difficulty in playing live with human performers. The synchronization function is critical for the interaction. We classify synchronization and musical expression into two levels: (1) the rhythm level and (2) the melody level. Two issues in achieving two-layer synchronization and musical expression are: (1) simultaneous estimation of the rhythm structure and the current part of the music and (2) derivation of the estimation confidence to switch behavior between the rhythm level and the melody level. This paper presents a score following algorithm, incremental audio to score alignment, that conforms to the two-level synchronization design using a particle filter. Our method estimates the score position for the melody level and the tempo for the rhythm level. The reliability of the score position estimation is extracted from the probability distribution of the score position. Experiments are carried out using polyphonic jazz songs. The results confirm that our method switches levels in accordance with the difficulty of the score estimation. When the tempo of the music is less than 120 (beats per minute; bpm), the estimated score positions are accurate and reported; when the tempo is over 120 (bpm), the system tends to report only the tempo to suppress the error in the reported score position predictions. Copyright © 2010, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.

  • Automatic allocation of training data for rapid prototyping of speech understanding based on multiple model combination

    Kazunori Komatani, Masaki Katsumaru, Mikio Nakano, Kotaro Funakoshi, Tetsuya Ogata, Hiroshi G. Okuno

    Coling 2010 - 23rd International Conference on Computational Linguistics, Proceedings of the Conference   2   579 - 587  2010年

     概要を見る

    The optimal choice of speech understanding method depends on the amount of training data available in rapid prototyping. A statistical method is ultimately chosen, but it is not clear at which point in the increase in training data a statistical method become effective. Our framework combines multiple automatic speech recognition (ASR) and language understanding (LU) modules to provide a set of speech understanding results and selects the best result among them. The issue is how to allocate training data to statistical modules and the selection module in order to avoid overfitting in training and obtain better performance. This paper presents an automatic training data allocation method that is based on the change in the coefficients of the logistic regression functions used in the selection module. Experimental evaluation showed that our allocation method outperformed baseline methods that use a single ASR module and a single LU module at every point while training data increase.

  • Analyzing user utterances in barge-in-able spoken dialogue system for improving identification accuracy

    Kyoko Matsuyama, Kazunori Komatani, Ryu Takeda, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010     3050 - 3053  2010年  [査読有り]

     概要を見る

    In our barge-in-able spoken dialogue system, the user's behaviors such as barge-in timing and utterance expressions vary according to his/her characteristics and situations. The system adapts to the behaviors by modeling them. We analyzed 1584 utterances collected by our systems of quiz and news-listing tasks and showed that ratio of using referential expressions depends on individual users and average lengths of listed items. This tendency was incorporated as a prior probability into our method and improved the identification accuracy of the user's intended items. © 2010 ISCA.

  • Effects of modelling within- and between-frame temporal variations in power spectra on non-verbal sound recognition

    Nobuhide Yamakawa, Tetsuro Kitahara, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 11th Annual Conference of the International Speech Communication Association, INTERSPEECH 2010     2342 - 2345  2010年  [査読有り]

     概要を見る

    Research on environmental sound recognition has not shown great development in comparison with that on speech and musical signals. One of the reasons is that the sound category of environmental sounds covers a broad range of acoustical natures. We classified them in order to explore suitable recognition techniques for each characteristic. We focus on impulsive sounds and their non-stationary feature within and between analytic frames. We used matching-pursuit as a framework to use wavelet analysis for extracting temporal variation of audio features inside a frame. We also investigated the validity of modeling decaying patterns of sounds using Hidden markov models. Experimental results indicate that sounds with multiple impulsive signals are recognized better by using time-frequency analyzing bases than by frequency domain analysis. Classification of sound classes with a long and clear decaying pattern improves when HMMs with multiple number of hidden states are applied. © 2010 ISCA.

  • SpeakBySinging: Converting singing voices to speaking voices while retaining voice timbre

    Shimpei Aso, Takeshi Saitou, Masataka Goto, Katsutoshi Itoyama, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    13th International Conference on Digital Audio Effects, DAFx 2010 Proceedings     14 - 121  2010年

     概要を見る

    This paper describes a singing-to-speaking synthesis system called "SpeakBySinging" that can synthesize a speaking voice from an input singing voice and the song lyrics. The system controls three acoustic features that determine the difference between speaking and singing voices: the fundamental frequency (F0), phoneme duration, and power (volume). By changing these features of a singing voice, the system synthesizes a speaking voice while retaining the timbre of the singing voice. The system first analyzes the singing voice to extract the F0 contour, the duration of each phoneme of the lyrics, and the power. These features are then converted to target values that are obtained by feeding the lyrics into a traditional text-to-speech (TTS) system. The system finally generates a speaking voice that preserves the timbre of the singing voice but has speech-like features. Experimental results show that SpeakBySinging can convert singing voices into speaking voices whose timbre is almost the same as the original singing voices.

  • Human-robot cooperation in arrangement of objects using confidence measure of neuro-dynamical system

    Hiromitsu Awano, Tetsuya Ogata, Shun Nishide, Toru Takahashi, Kazunori Komatani, Hiroshi G. Okuno

    Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics     2533 - 2538  2010年  [査読有り]

     概要を見る

    The objective of our study was to develop dynamic collaboration between a human and a robot. Most conventional studies have created pre-designed rule-based collaboration systems to determine the timing and behavior of robots to participate in tasks. Our aim is to introduce the confidence of the task as a criterion for robots to determine their timing and behavior. In this paper, we report the effectiveness of applying reproduction accuracy as a measure for quantitatively evaluating confidence in an object arrangement task. Our method is comprised of three phases. First, we obtain human-robot interaction data through the Wizard of OZ method. Second, the obtained data are trained using a neuro-dynamical system, namely, the Multiple Time-scales Recurrent Neural Network (MTRNN). Finally, the prediction error in MTRNN is applied as a confidence measure to determine the robot's behavior. The robot participated in the task when its confidence was high, while it just observed when its confidence was low. Training data were acquired using an actual robot platform, Hiro. The method was evaluated using a robot simulator. The results revealed that motion trajectories could be precisely reproduced with a high degree of confidence, demonstrating the effectiveness of the method. ©2010 IEEE.

    DOI

  • Exploiting harmonic structures to improve separating simultaneous speech in under-determined conditions

    Yasuharu Hirasawa, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings     450 - 457  2010年  [査読有り]

     概要を見る

    In real-world situations, a robot may often encounter "under- determined" situation, where there are more sound sources than microphones. This paper presents a speech separation method using a new constraint on the harmonic structure for a simultaneous speech-recognition system in under-determined conditions. The requirements for a speech separation method in a simultaneous speech-recognition system are (1) ability to handle a large number of talkers, and (2) reduction of distortion in acoustic features. Conventional methods use a maximum likelihood estimation in sound source separation, which fulfills requirement (1). Since it is a general approach, the performance is limited when separating speech. This paper presents a two-stage method to improve the separation. The first stage uses maximum likelihood estimation and extracts the harmonic structure, and the second stage exploits the harmonic structure as a new constraint to achieve requirement (2). We carried out an experiment that simulated three simultaneous utterances using impulse responses recorded by two microphones in an anechoic chamber. The experimental results revealed that our method could improve speech recognition correctness by about four points. ©2010 IEEE.

    DOI

  • Motion generation based on reliable predictability using self-organized object features

    Shun Nishide, Tetsuya Ogata, Jun Tani, Toru Takahashi, Kazunori Komatani, Hiroshi G. Okuno

    IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings     3453 - 3458  2010年  [査読有り]

     概要を見る

    Predictability is an important factor for determining robot motions. This paper presents a model to generate robot motions based on reliable predictability evaluated through a dynamics learning model which self-organizes object features. The model is composed of a dynamics learning module, namely Recurrent Neural Network with Parametric Bias (RNNPB), and a hierarchical neural network as a feature extraction module. The model inputs raw object images and robot motions. Through bi-directional training of the two models, object features which describe the object motion are self-organized in the output of the hierarchical neural network, which is linked to the input of RNNPB. After training, the model searches for the robot motion with high reliable predictability of object motion. Experiments were performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing, and rolling motions. For objects with single motion possibility, the robot tended to generate motions that induce the object motion. For objects with two motion possibilities, the robot evenly generated motions that induce the two object motions. ©2010 IEEE.

    DOI

  • An improvement in automatic speech recognition using soft missing feature masks for robot audition

    Toru Takahashi, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings     964 - 969  2010年  [査読有り]

     概要を見る

    We describe integration of preprocessing and automatic speech recognition based on Missing-Feature-Theory (MFT) to recognize a highly interfered speech signal, such as the signal in a narrow angle between a desired and interfered speakers. As a speech signal separated from a mixture of speech signals includes the leakage from other speech signals, recognition performance of the separated speech degrades. An important problem is estimating the leakage in time-frequency components. Once the leakage is estimated, we can generate missing feature masks (MFM) automatically by using our method. A new weighted sigmoid function is introduced for our MFM generation method. An experiment shows that a word correct rate improves from 66 % to 74 % by using our MFM generation method tuned by a search base approach in the parameter space. ©2010 IEEE.

    DOI

  • Robot musical accompaniment: Integrating audio and visual cues for real-time synchronization with a human flutist

    Angelica Lim, Takeshi Mizumoto, Louis Kenzo Cahier, Takuma Otsuka, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings     1964 - 1969  2010年  [査読有り]

     概要を見る

    Musicians often have the following problem: they have a music score that requires 2 or more players, but they have no one with whom to practice. So far, score-playing music robots exist, but they lack adaptive abilities to synchronize with fellow players' tempo variations. In other words, if the human speeds up their play, the robot should also increase its speed. However, computer accompaniment systems allow exactly this kind of adaptive ability. We present a first step towards giving these accompaniment abilities to a music robot. We introduce a new paradigm of beat tracking using 2 types of sensory input - visual and audio - using our own visual cue recognition system and state-of-the-art acoustic onset detection techniques. Preliminary experiments suggest that by coupling these two modalities, a robot accompanist can start and stop a performance in synchrony with a flutist, and detect tempo changes within half a second. ©2010 IEEE.

    DOI

  • Human-robot ensemble between robot thereminist and human percussionist using coupled oscillator model

    Takeshi Mizumoto, Takuma Otsuka, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings     1957 - 1963  2010年  [査読有り]

     概要を見る

    This paper presents a novel synchronizing method for a human-robot ensemble using coupled oscillators. We define an ensemble as a synchronized performance produced through interactions between independent players. To attain better synchronized performance, the robot should predict the human's behavior to reduce the difference between the human's and robot's onset timings. Existing studies in such synchronization only adapts to onset intervals, thus, need a considerable time to synchronize. We use a coupled oscillator model to predict the human's behavior. Experimental results show that our method reduces the average of onset time errors; when we use a metronome, a tempo-varying metronome or a human drummer, errors are reduced by 38%, 10% or 14% on the average, respectively. These results mean that the prediction of human's behaviors is effective for the synchronized performance. ©2010 IEEE.

    DOI

  • Speedup and performance improvement of ica-based robot audition by parallel and resampling-based block-wise processing

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE/RSJ 2010 International Conference on Intelligent Robots and Systems, IROS 2010 - Conference Proceedings     1949 - 1956  2010年  [査読有り]

     概要を見る

    This paper describes a speedup and performance improvement of multi-channel semi-blind ICA (MCSB-ICA) with parallel and resampling-based block-wise processing. MCSB-ICA is an integrated method of sound source separation that accomplishes blind source separation, blind dereverberation, and echo cancellation. This method enables robots to separate user's speech signals from observed signals including the robot's own speech, other speech and their reverberations without a priori information. The main problem when MCSB-ICA is applied to robot audition is its high computational cost. We tackle this by multi-threading programming, and the two main issues are 1) the design of parallel processing and 2) incremental implementation. These are solved by a) multiple-stack-based parallel implementation, and b) resampling-based overlaps and block-wise separation. The experimental results proved that our method reduced the real-time factor to less than 0.5 with an eight-core CPU, and it improves the performance of automatic speech recognition by 2.10 points compared with the single-stack-based parallel implementation without the resampling technique. ©2010 IEEE.

    DOI

  • Method of discriminating known and unknown environmental sounds using recurrent neural network

    Yang Zhang, Tetsuya Ogata, Shun Nishide, Toru Takahashi, Hiroshi G. Okuno

    SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems     378 - 383  2010年

     概要を見る

    This paper describes our method to classify nonspeech environmental sounds for robots working. In the real world, two main restrictions pertain in learning. First, robots have to learn using only a small amount of sounds in a limited time and space because of restrictions. Second, it has to detect unknown sounds to avoid false classification since it is virtually impossible to collect samples of all environmental sounds. Most of the previous methods require a huge number of samples of all target sounds, including noises, for training stochastic models such as the Gaussian mixture model. In contrast, we use a neurodynamical model to build a prediction and classification system. The neuro-dynamical system can be trained with a small amount of sounds and generalize others by inferring the sound generation dynamics. After training, a self-organized space is structured for the sound generation dynamics. The proposed system classify on the basis of the self-organized space. The prediction results of sounds are used for determining unknown sounds in our system. In this paper, we show the results of preliminary experiments on the proposed model's classification of known and unknown sound classes.

  • Query-by-example music information retrieval by score-informed source separation and remixing technologies

    Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Eurasip Journal on Advances in Signal Processing   2010  2010年  [査読有り]

     概要を見る

    We describe a novel query-by-example (QBE) approach in music information retrieval that allows a user to customize query examples by directly modifying the volume of different instrument parts. The underlying hypothesis of this approach is that the musical mood of retrieved results changes in relation to the volume balance of different instruments. On the basis of this hypothesis, we aim to clarify the relationship between the change in the volume balance of a query and the genre of the retrieved pieces, called genre classification shift. Such an understanding would allow us to instruct users in how to generate alternative queries without finding other appropriate pieces. Our QBE system first separates all instrument parts from the audio signal of a piece with the help of its musical score, and then it allows users remix these parts to change the acoustic features that represent the musical mood of the piece. Experimental results showed that the genre classification shift was actually caused by the volume change in the vocal, guitar, and drum parts. © 2010 Katsutoshi Itoyama et al.

    DOI

  • Missing-feature-theory-based robust simultaneous speech recognition system with non-clean speech acoustic model

    Toru Takahashi, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009     2730 - 2735  2009年12月

     概要を見る

    A humanoid robot must recognize a target speech signal while people around the robot chat with them in real-world. To recognize the target speech signal, robot has to separate the target speech signal among other speech signals and recognize the separated speech signal. As separated signal includes distortion, automatic speech recognition (ASR) performance degrades. To avoid the degradation, we trained an acoustic model from non-clean speech signals to adapt acoustic feature of distorted signal and adding white noise to separated speech signal before extracting acoustic feature. The issues are (1) To determine optimal noise level to add the training speech signals, and (2) To determine optimal noise level to add the separated signal. In this paper, we investigate how much noises should be added to clean speech data for training and how speech recognition performance improves for different positions of three talkers with soft masking. Experimental results show that the best performance is obtained by adding white noises of 30 dB. The ASR with the acoustic model outperforms with ASR with the clean acoustic model by 4 points. © 2009 IEEE.

    DOI

  • Incremental polyphonic audio to score alignment using beat tracking for singer robots

    Takuma Otsuka, Kazumasa Murata, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009     2289 - 2296  2009年12月  [査読有り]

     概要を見る

    We aim at developing a singer robot capable of listening to music with its own "ears" and interacting with a human's musical performance. Such a singer robot requires at least three functions: listening to the music, understanding what position in the music is being performed, and generating a singing voice. In this paper, we focus on the second function, that is, the capability to align an audio signal to its musical score represented symbolically. Issues underlying the score alignment problem are: (1) diversity in the sounds of various musical instruments, (2) difference between the audio signal and the musical score, (3) fluctuation in tempo of the musical performance. Our solutions to these issues are as follows: (1) the design of features based on a chroma vector in the 12-tone model and onset of the sound, (2) defining the rareness for each tone based on the idea that scarcely used tone is salient in the audio signal, and (3) the use of a switching Kalman filter for robust tempo estimation. The experimental result shows that our score alignment method improves the average of cumulative absolute errors in score alignment by 29% using 100 popular music tunes compared to the beat tracking without score alignment. © 2009 IEEE.

    DOI

  • Phoneme acquisition model based on vowel imitation using recurrent neural network

    Hisashi Kanda, Tetsuya Ogata, Toru Takahashi, Kazunori Komatani, Hiroshi G. Okuno

    2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009     5388 - 5393  2009年12月  [査読有り]

     概要を見る

    A phoneme-acquisition system was developed using a computational model that explains the developmental process of human infants in the early period of acquiring language. There are two important findings in constructing an infant's acquisition of phonemes: (1) an infant's vowel like cooing tends to invoke utterances that are imitated by its caregiver, and (2) maternal imitation effectively reinforces infant vocalization. Therefore, we hypothesized that infants can acquire phonemes to imitate their caregivers' voices by trial and error, i. e., infants use self-vocalization experience to search for imitable and unimitable elements in their caregivers' voices. On the basis of this hypothesis, we constructed a phoneme acquisition process using interaction involving vowel imitation between a human and an infant model. Our infant model had a vocal tract system, called the Maeda model, and an auditory system implemented by using Mel-Frequency Cepstral Coefficients (MFCCs) through STRAIGHT analysis. We applied Recurrent Neural Network with Parametric Bias (RNNPB) to learn the experience of self-vocalization, to recognize the human voice, and to produce the sound imitated by the infant model. To evaluate imitable and unimitable sounds, we used the prediction error of the RNNPB model. The experimental results revealed that as imitation interactions were repeated, the formants of sounds imitated by our system moved closer to those of human voices, and our system could self-organize the same vowels in different continuous sounds. This suggests that our system can reflect the process of phoneme acquisition. © 2009 IEEE.

    DOI

  • Emergence of evolutionary interaction with voice and motion between two robots using RNN

    Wataru Hinoshita, Tetsuya Ogata, Hideki Kozima, Hisashi Kanda, Toru Takahashi, Hiroshi G. Okuno

    2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009     4186 - 4192  2009年12月  [査読有り]

     概要を見る

    We propose a model of evolutionary interaction between two robots where signs used for communication emerge through mutual adaptation. Signs used in human interaction, e.g., language, gestures and eye contact change and evolve in form and meaning through repeated use. To create flexible human-like interaction systems, it is necessary to deal with signs as a dynamic property and to construct a framework in which signs emerge from mutual adaptation by agents. Our target is multi-modal interaction using voice and motion between two robots where a voice/motion pattern is used as a sign referring to a motion/voice pattern. To enable evolutionary signs (voice and motion patterns) to be recognized and generated, we utilized a dynamics model: Multiple Timescale Recurrent Neural Network (MTRNN). To enable the robots to interpret signs, we utilized hierarchical neural networks, which transform dynamics model parameters of voice/motion into those of motion/voice. In our experiment, two robots modified their own interpretation of signs constantly through mutual adaptation in interaction where they responded to the other's voice with motion one after the other. As a result of the experiment, we found that the interaction kept evolving through the robots' repeated and alternate miscommunications and readaptations, and this induced the emergence of diverse new signs that depended on the robots' body dynamics through the generalization capability of MTRNN. © 2009 IEEE.

    DOI

  • Step-size parameter adaptation of multi-channel semi-blind ICA with piecewise linear model for barge-in-able robot audition

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009     2277 - 2282  2009年12月  [査読有り]

     概要を見る

    This paper describes a step-size parameter adaptation technique of multi-channel semi-blind independent component analysis (MCSB-ICA) for a "barge-in-able" robot audition system. By "barge-in", we mean that the user can speak simultaneously when the robot is speaking. We focused on MCSB-ICA to achieve such an audition system because it can separate a user's and a robot's speech under reverberant environments. The problem with MCSB-ICA for robot audition is the slow speed of convergence in estimating a separation filter due to its step-size parameters. Many optimization methods cannot be adopted because their computational costs are proportional to the 2nd order of the reverberation time. Our method yields adaptive step-size parameters with MCSB-ICA at low computational costs. It is based on three techniques; 1) recursive expression of the separation process, 2) a piecewise linear model of the step-size of the separation filter, and 3) adaptive step-size parameters with a sub-ICA-filter. Experimental results show that our approach attains faster convergence speed and lower computational costs than those with a fixed step-size parameter. © 2009 IEEE.

    DOI

  • Modeling tool-body assimilation using second-order recurrent neural network

    Shun Nishide, Tatsuhiro Nakagawa, Tetsuya Ogata, Jun Tani, Toru Takahashi, Hiroshi G. Okuno

    2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2009     5376 - 5381  2009年12月  [査読有り]

     概要を見る

    Tool-body assimilation is one of the intelligent human abilities. Through trial and experience, humans are capable of using tools as if they are part of their own bodies. This paper presents a method to apply a robot's active sensing experience for creating the tool-body assimilation model. The model is composed of a feature extraction module, dynamics learning module, and a tool recognition module. Self-Organizing Map (SOM) is used for the feature extraction module to extract object features from raw images. Multiple Time-scales Recurrent Neural Network (MTRNN) is used as the dynamics learning module. Parametric Bias (PB) nodes are attached to the weights of MTRNN as second-order network to modulate the behavior of MTRNN based on the tool. The generalization capability of neural networks provide the model the ability to deal with unknown tools. Experiments are performed with HRP-2 using no tool, I-shaped, T-shaped, and L-shaped tools. The distribution of PB values have shown that the model has learned that the robot's dynamic properties change when holding a tool. The results of the experiment show that the tool-body assimilation model is capable of applying to unknown objects to generate goal-oriented motions. © 2009 IEEE.

    DOI

  • Thereminist Robot: Development of a Robot Theremin Player based on a Theremin”s Pitch Model

    Takeshi Mizumoto, Hiroshi Tsujino, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Special SIGMUS Symposium     2297 - 2302  2009年12月  [査読有り]

    DOI

  • 弦の音響差異を考慮したバイオリン演奏音響信号に対する運指推定

    前澤陽, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    日本音響学会関西支部第12回若手研究者交流研究発表会    2009年12月

  • ハードウェアをカスタマイズできるコミュニケーションロボットに関する研究

    守良真, 由美子, 近藤裕樹, 菅佑樹, 尾形哲也, 菅野重樹

    第10回システムインテグレーション部門講演会 (SI2009), 計測自動制御学会     2F1 - 2  2009年12月

  • MTRNNによる環境音の予測識別

    張陽, 尾形哲也, 高橋徹, 駒谷和範, 奥乃博

    第10回システムインテグレーション部門講演会 (SI2009), 計測自動制御学会     1H4 - 5  2009年12月

  • Self-organization of dynamic object features based on bidirectional training

    Shun Nishide, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    Advanced Robotics   23 ( 15 ) 2035 - 2057  2009年10月  [査読有り]

     概要を見る

    This paper presents a method to self-organize object features that describe object dynamics using bidirectional training. The model is composed of a dynamics learning module and a feature extraction module. Recurrent Neural Network with Parametric Bias (RNNPB) is utilized for the dynamics learning module, learning and self-organizing the sequences of robot and object motions. A hierarchical neural network is linked to the input of RNNPB as the feature extraction module for self-organizing object features that describe the object motions. The two modules are simultaneously trained through bidirectional training using image and motion sequences acquired from the robot's active sensing with objects. Experiments are performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing and rolling motions. The results have shown that the model is capable of self-organizing object dynamics based on the self-organized features. © Koninklijke Brill NV, Leiden and The Robotics Society of Japan, 2009.

    DOI

  • Target speech detection and separation for communication with humanoid robots in noisy home environments

    Hyun Don Kim, Jinsung Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Advanced Robotics   23 ( 15 ) 2093 - 2111  2009年10月  [査読有り]

     概要を見る

    People usually talk face to face when they communicate with their partner. Therefore, in robot audition, the recognition of the front talker is critical for smooth interactions. This paper presents an enhanced speech detection method for a humanoid robot that can separate and recognize speech signals originating from the front even in noisy home environments. The robot audition system consists of a new type of voice activity detection (VAD) based on the complex spectrum circle centroid (CSCC) method and a maximum signal-to-noise ratio (SNR) beamformer. This VAD based on CSCC can classify speech signals that are retrieved at the frontal region of two microphones embedded on the robot. The system works in real-time without needing training filter coefficients given in advance even in a noisy environment (SNR > 0 dB). It can cope with speech noise generated from televisions and audio devices that does not originate from the center. Experiments using a humanoid robot, SIG2, with two microphones showed that our system enhanced extracted target speech signals more than 12 dB (SNR) and the success rate of automatic speech recognition for Japanese words was increased by about 17 points. © Koninklijke Brill NV, Leiden and The Robotics Society of Japan, 2009.

    DOI

  • Enabling A User To Specify An Item At Any Time During System Enumeration

    Kyoko MATSUYAMA, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of International Conference on Spoken Language Processing (Interspeech-2009)     4 - 1  2009年09月

  • RNNを備えた2体のロボット間における身体性に基づいた動的コミュニケーションモデル

    日下航, 尾形哲也, 小嶋秀樹, 高橋徹, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • 二次リカレントニューラルネットワークを用いた道具身体化モデルの構築

    西出俊, 中川達裕, 尾形哲也, 谷淳, 高橋徹, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • ロボット音声対話におけるバージン発話の指示対象同定

    松山匡子, 駒谷和範, 武田龍, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • テルミン演奏ロボットのための音高依存性を考慮した音量モデル

    水本武志, 辻野広司, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • Voice-Awareness Control Consistent with Robot”s Body Movements

    大塚琢馬, 中臺一博, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • 頭部音響伝達関数を用いたGSSによる3話者同時発話認識?HARK 1.0.0 の新機能?

    高橋徹, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • 声道物理モデルの母音列繰り返し模倣による音素獲得シミュレーション (基調講演)

    尾形哲也, 神田尚, 高橋徹, 駒谷和範, 奥乃博

    日本ロボット学会第27回学術講演会, 横浜国立大学    2009年09月

  • Binaural active audition for humanoid robots to localise speech over entire azimuth range

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, HIROShi G. Okuno

    Applied Bionics and Biomechanics   6 ( 3-4 ) 355 - 367  2009年09月

     概要を見る

    We applied motion theory to robot audition to improve the inadequate performance. Motions are critical for overcoming the ambiguity and sparseness of information obtained by two microphones. To realise this, we first designed a sound source localisation system integrated with cross-power spectrum phase (CSP) analysis and an EM algorithm. The CSP of sound signals obtained with only two microphones was used to localise the sound source without having to measure impulse response data. The expectation-maximisation (EM) algorithm helped the system to cope with several moving sound sources and reduce localisation errors. We then proposed a way of constructing a database for moving sounds to evaluate binaural sound source localisation. We evaluated our sound localisation method using artificial moving sounds and confirmed that it could effectively localise moving sounds slower than 1.125 rad/s. Consequently, we solved the problem of distinguishing whether sounds were coming from the front or rear by rotating and/or tipping the robot's head that was equipped with only two microphones. Our system was applied to a humanoid robot called SIG2, and we confirmed its ability to localise sounds over the entire azimuth range as the success rates for sound localisation in the front and rear areas were 97.6% and 75.6% respectively. © 2009 Taylor & Francis.

    DOI

  • 音声対話システムにおける文法検証結果と発話履歴に基づくヘルプメッセージ候補のランキング

    駒谷和範, 池田智志, 福林雄一朗, 尾形哲也, 奥乃博

    情報処理学会音声言語情報処理研究会, 飯坂温泉,情報処理学会.    2009年07月

  • 音響信号と音楽的制約を統合したバイオリンの演奏弦系列の推定

    前澤陽, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    音楽情報科学研究会, 情報処理学会   Vol.2009-MUS-81 ( No.5 ) 1 - 6  2009年07月

  • 複数楽器混合モデルのパラメータ推定と楽器名同定への応用

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報科学研究会, 情報処理学会   Vol.2009-MUS-81 ( No.13 ) 1 - 6  2009年07月

  • 残差スペクトルモデルによる伴奏・残響成分抑制に基づいた楽器演奏分析合成の高精度化

    安良岡直希, 安部武宏, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    音楽情報科学研究会   Vol.2009-MUS-81 ( No.10 ) 1 - 6  2009年07月

  • 多重奏楽曲の楽器音量バランス変化による音楽ジャンルシフト

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報科学研究会   Vol.2009-MUS-8 ( No.3 ) 1 - 6  2009年07月

  • 市民参画のための公的討議の議事録閲覧支援システム

    白松俊, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会全国大会 (JSAI2009)     3I1 - 1  2009年06月

  • バージイン発話タイミングを導入した指示対象同定

    松山匡子, 駒谷和範, 武田龍, 尾形哲也, 奥乃博

    情報処理学会音声言語研究会    2009年05月

  • 歌唱ロボットのためのビート情報とメロディ・ハーモニー情報の統合による音楽音響信号と楽譜の実時間同期手法の開発

    大塚 琢馬, 村田 和真, 武田 龍, 中臺 一博, 高橋 徹, 尾形 哲也, 奥乃 博

    全国大会講演論文集   71 ( 0 ) 243 - 244  2009年03月

    CiNii

  • 音色特徴の歪みを回避した楽器音の音高・音長操作手法

    安部武宏, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   50 ( 3 ) 1054 - 1066  2009年03月

  • 話題遷移図の可視化と話題遷移に応じた関連情報提示による議事録閲覧支援

    白松俊, 駒谷和範, 尾形哲也, 高橋徹, 奥乃博

    言語処理学会第15回年次大会     D2 - 1  2009年03月

  • RNNを備えた2対の小型ロボット間の首振り動作と音声によるインタラクションにおける共有シンボルの創発

    日下航, 神田尚, 尾形哲也, 小嶋秀樹, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 神経回路モデルを用いた音声模倣モデルによる音声バブリングと音声獲得過程シミュレーション

    神田尚, 尾形哲也, 高橋徹, 駒谷和範, 奥乃 博

    情報処理学会第71回全国大会    2009年03月

  • 実環境音声対話システムにおけるバージイン発話タイミングを活用した指示対象の同定

    松山匡子, 駒谷和範, 白松俊, 武田龍, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 音声認識と言語理解を動的に選択する音声理解フレームワーク

    勝丸真樹, 中野幹生, 駒谷和範, 成松宏美, 船腰孝太郎, 辻野広司, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 音声対話システムにおける想定外発話の文法検証を用いた対話行為推定に基づくヘルプ生成,

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 顔追跡による音環境可視化システムのアウエアネスの改善

    久保田祐史, 白松俊, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • フィールドにおける音源定位のための音声視覚化デバイス「カエルホタル」の設計

    水本武志, 合原一究, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • Probabilistic Classification of Monophonic Instrument Playing Techniques

    前澤陽, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 楽器の内部モデルに基づくフィードフォワード制御によるテルミン演奏ロボットの開発

    水本武志, 辻野広司, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • ロボットによる卓上物体操作のためのRNNを用いた道具身体化モデルの構築

    中川達裕, 尾形哲也, 谷淳, 高橋徹, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • ソフトマスクと音響モデル適応を用いた3話者同時発話音声認識

    高橋徹, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • A Music Retrieval Approach from Alternative Genres of Query by Adjusting Instrument Volume

    王凱平, 糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 連続発音中の音色変化に着目した未学習譜面情への演奏信号生成

    安良岡直希, 安部武宏, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • ベース音高とクロマベクトルの相関に基づいた和音進行認識

    高野秀樹, 須見康平, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • 音色特徴量に基づく調波・非調波統合モデルによる楽器音モーフィング

    安部武宏, 糸山克寿, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第71回全国大会    2009年03月

  • マルチドメイン音声対話システムにおけるトピック推定と対話履歴の統合によるドメイン選択手法

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   50 ( 2 ) 488 - 500  2009年02月

  • 複数の言語モデル・言語理解方式を用いた音声理解の高精度化

    勝丸真樹, 中野幹生, 駒谷和範, 成松宏美, 船腰孝太郎, 辻野広司, 高橋徹, 尾形哲也, 奥乃博

    第75回音声言語情報処理研究会, 2009-SLP-75 (9), 情処研報   Vol.2009 ( No.10 ) 45 - 50  2009年02月

  • 自己形態主張を行うカスタマイズ可能なコミュニケーションロボットの研究

    守良真, 近藤裕樹, 奥出京司郎, 菅佑樹, 尾形哲也, 菅野重樹

    日本機械学会ロボティクス・メカトロニクス部門(ROBOMEC2009)   1P1-F11  2009年

  • ハードウェアをカスタマイズできるコミュニケーションロボットにおける研究

    近藤裕樹, 守良真, 奥出京司郎, 菅佑樹, 尾形哲也, 菅野重樹

    日本機械学会ロボティクス・メカトロニクス部門(ROBOMEC2009)   1P1-E21  2009年

  • Enabling a user to specify an item at any time during system enumeration - Item identification for barge-in-able conversational dialogue systems

    Kyoko Matsuyama, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH     252 - 255  2009年  [査読有り]

     概要を見る

    In conversational dialogue systems, users prefer to speak at any time and to use natural expressions. We have developed an Independent Component Analysis (ICA) based semi-blind source separation method, which allows users to barge-in over system utterances at any time. We created a novel method from timing information derived from barge-in utterances to identify one item that a user indicates during system enumeration. First, we determine the timing distribution of user utterances containing referential expressions and then approximate it using a gamma distribution. Second, we represent both the utterance timing and automatic speech recognition (ASR) results as probabilities of the desired selection from the system's enumeration. We then integrate these two probabilities to identify the item having the maximum likelihood of selection. Experimental results using 400 utterances indicated that our method outperformed two methods used as a baseline (one of ASR results only and one of utterance timing only) in identification accuracy. Copyright © 2009 ISCA.

  • Ranking help message candidates based on robust grammar verification results and utterance history in spoken dialogue systems

    Kazunori Komatani, Satoshi Ikeda, Yuichiro Fukubayashi, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the SIGDIAL 2009 Conference: 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue     314 - 321  2009年  [査読有り]

     概要を見る

    We address an issue of out-of-grammar (OOG) utterances in spoken dialogue systems by generating help messages for novice users. Help generation for OOG utterances is a challenging problem because language understanding (LU) results based on automatic speech recognition (ASR) results for such utterances are always erroneous as important words are often misrecognized or missed from such utterances. We first develop grammar verification for OOG utterances on the basis of a Weighted Finite-State Transducer (WFST). It robustly identifies a grammar rule that a user intends to utter, even when some important words are missed from the ASR result. We then adopt a ranking algorithm, RankBoost, whose features include the grammar verification results and the utterance history representing the user's experience. © 2009 Association for Computational Linguistics.

    DOI

  • A Speech Understanding Framework that Uses Multiple Language Models and Multiple Understanding Models.

    Masaki Katsumaru, Mikio Nakano, Kazunori Komatani, Kotaro Funakoshi, Tetsuya Ogata, Hiroshi G. Okuno

    Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics, Proceedings, May 31 - June 5, 2009, Boulder, Colorado, USA, Short Papers     133 - 136  2009年  [査読有り]

  • 残響下でのバージイン発話認識のための多入力独立成分分析を応用したロボット聴覚

    武田龍, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会誌   27 ( 7 ) 782 - 792  2009年

  • 人工神経回路モデルと声道物理モデルを用いた母音模倣モデルに基づく音素獲得シミュレーション

    神田尚, 尾形哲也, 高橋徹, 駒谷和範, 奥乃博

    日本ロボット学会誌   27 ( 7 ) 802 - 813  2009年

  • Autonomous Motion Generation Based on Reliable Predictability.

    Shun Nishide, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    J. Robotics Mechatronics   21 ( 4 ) 478 - 488  2009年

    DOI

  • Adjusting occurrence probabilities of automatically-generated abbreviated words in spoken dialogue systems

    Masaki Katsumaru, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   5579 LNAI   481 - 490  2009年  [査読有り]

     概要を見る

    Users often abbreviate long words when using spoken dialogue systems, which results in automatic speech recognition (ASR) errors. We define abbreviated words as sub-words of an original word and add them to the ASR dictionary. The first problem we face is that proper nouns cannot be correctly segmented by general morphological analyzers, although long and compound words need to be segmented in agglutinative languages such as Japanese. The second is that, as vocabulary size increases, adding many abbreviated words degrades the ASR accuracy. We have developed two methods, (1) to segment words by using conjunction probabilities between characters, and (2) to adjust occurrence probabilities of generated abbreviated words on the basis of the following two cues: phonological similarities between the abbreviated and original words and frequencies of abbreviated words in Web documents. Our method improves ASR accuracy by 34.9 points for utterances containing abbreviated words without degrading the accuracy for utterances containing original words. © 2009 Springer Berlin Heidelberg Spoken dialogue systems*abbreviated words*adjusting occurrence probabilities.

    DOI

  • Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09     250 - 255  2009年

     概要を見る

    This paper presents an ICA-based robot audition system which estimates the reverberation time of the environment automatically by using the robot's own speech. The system is based on multi-channel semi-blind independent component analysis (MCSB-ICA), a source separation method using a microphone array that can separate user and robot speech under reverberant environments. Perception of the reverberation time (RT) is critical, because an inappropriate RT degrades separation performance and increases processing time. Unlike most previous methods that assume the RT is given in advance, our method estimates an RT by using the echo's intensity of the robot's own speech. It has three steps: speaks a sentence in a new environment, calculates the relative powers of the echoes, and estimates the RT using linear regression of them. Experimental results show that this method sets an appropriate RT for MCSB-ICA for real-world environments and that word correctness is improved by up to 6 points and processing time is reduced by up to 60%. ©2009 IEEE.

    DOI

  • ICA-based efficient blind dereverberation and echo cancellation method for barge-in-able robot audition

    Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings     3677 - 3680  2009年  [査読有り]

     概要を見る

    This paper describes a new method that allows "Barge-In" in various environments for robot audition. "Barge-in" means that a user begins to speak simultaneously while a robot is speaking. To achieve the function, we must deal with problems on blind dereverberation and echo cancellation at the same time. We adopt Independent Component Analysis (ICA) because it essentially provides a natural framework for these two problems. To deal with reverberation, we apply a Multiple Input/Output INverse-filtering Theorem-based model of observation to the frequency domain ICA. The main problem is its high-computational cost of ICA. We reduce the computational complexity to the linear order of reverberation time by using two techniques: 1) a separation model based on observed signal independence, and 2) enforced spatial sphering for preprocessing. The experimental results revealed that our method improved word correctness of reverberant speech by 10-20 points. ©2009 IEEE.

    DOI

  • Human tracking system integrating sound and face localization using an expectation-maximization algorithm in real environments

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Advanced Robotics   23 ( 6 ) 629 - 653  2009年  [査読有り]

     概要を見る

    We have developed a human tracking system for use by robots that integrate sound and face localization. Conventional systems usually require many microphones and/or prior information to localize several sound sources. Moreover, they are incapable of coping with various types of background noise. Our system, the cross-power spectrum phase analysis of sound signals obtained with only two microphones, is used to localize the sound source without having to use prior information such as impulse response data. An expectation- maximization (EM) algorithm is used to help the system cope with several moving sound sources. The problem of distinguishing whether sounds are coming from the front or back is also solved with only two microphones by rotating the robot's head. A developed method that uses facial skin colors classified by another EM algorithm enables the system to detect faces in various poses. It can compensate for the error in the sound localization for a speaker and also identify noise signals entering from undesired directions by detecting a human face. A developed probability-based method is used to integrate the auditory and visual information in order to produce a reliable tracking path in real-time. Experiments using a robot showed that our system can localize two sounds at the same time and track a communication partner while dealing with various types of background noise. © 2009 Koninklijke Brill NV.

    DOI

  • Continuous vocal imitation with self-organized vowel spaces in recurrent neural network

    Hisashi Kanda, Tetsuya Ogata, Toru Takahashi, Kazunori Komatani, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     4438 - 4443  2009年  [査読有り]

     概要を見る

    A continuous vocal imitation system was developed using a computational model that explains the process of phoneme acquisition by infants. Human infants perceive speech sounds not as discrete phoneme sequences but as continuous acoustic signals. One of critical problems in phoneme acquisition is the design for segmenting these continuous speech sounds. The key idea to solve this problem is that articulatory mechanisms such as the vocal tract help human beings to perceive speech sound units corresponding to phonemes. To segment acoustic signal with articulatory movement, we apply the segmenting method to our system by Recurrent Neural Network with Parametric Bias (RNNPB). This method determines the multiple segmentation boundaries in a temporal sequence using the prediction error of the RNNPB model, and the PB values obtained by the method can be encoded as kind of phonemes. Our system was implemented by using a physical vocal tract model, called the Maeda model. Experimental results demonstrated that our system can self-organize the same phonemes in different continuous sounds, and can imitate vocal sound involving arbitrary numbers of vowels using the vowel space in the RNNPB. This suggests that our model reflects theprocess of phoneme acquisition. © 2009 IEEE.

    DOI

  • Prediction and Imitation of Other”s Motions by Reusing Own Forward-Inverse Model in Robots

    Tetsuya OGATA, Ryunosuke YOKOYA, Jun TANI, Kazunori KOMATANI, Hiroshi G. OKUNO

    Proceedings of IEEE-RAS International Conference on Robots and Automation (ICRA-2009)     4144 - 4149  2009年  [査読有り]

    DOI

  • Visualization-based approaches to support context sharing towards public involvement support system

    Shun Shiramatsu, Yuji Kubota, Kazunori Komatani, Tetsuya Ogata, Toru Takahashi, Hiroshi G. Okuno

    Studies in Computational Intelligence   214   111 - 117  2009年  [査読有り]

     概要を見る

    In order to facilitate public involvement in the consensus building process needed for community development, a lot of time and effort needs to be spent on assessing and sharing public concerns. This paper presents new approaches for support for context sharing that involve visualizing public meeting records. The first approach is to visualize the transition of topics to enable the user to grasp an overview and to find specific arguments. The second is to visualize topic-related information to enable the user to understand background. The third is to visualize the auditory scene to enable the user to find and to listen to paralinguistic (prosodic) information contained in audio recordings. These approaches are designed on the basis of Visual Information-Seeking Mantra, "Overview first, zoom and filter, then details on demand." These approaches support citizens and stakeholders to find, to track, and to understand target arguments from the records of a public meeting. © 2009 Springer-Verlag Berlin Heidelberg.

    DOI

  • Parameter estimation for harmonic and inharmonic models by using timbre feature distributions

    Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Journal of Information Processing   17 ( 7 ) 191 - 201  2009年

     概要を見る

    We describe an improved way of estimating parameters for an integrated weighted-mixture model consisting of both harmonic and inharmonic tone models. Our final goal is to build an instrument equalizer (music remixer) that enables a user to change the volume of parts of polyphonic sound mixtures. To realize the instrument equalizer, musical signals must be separated into each musical instrument part. We have developed a score-informed sound source separation method using the integrated model. A remaining but critical problem is to find a way to deal with timbre varieties caused by various performance styles and instrument bodies because our method used template sounds to represent their timbre. Template sounds are generated from a MIDI tone generator based on an aligned score. Difference of instrument bodies between mixed signals and template sounds causes timbre difference and decreases separation performance. To solve this problem, we train probabilistic distributions of timbre features using various sounds to reduce template dependency. By adding a new constraint of maximizing the likelihood of timbre features extracted from each tone model, we can estimate model parameters that express the timbre more accurately. Experimental results show that separation performance improved from 4.89 to 8.48 dB.

    DOI

  • Analysis of motion searching based on reliable predictability using recurrent neural network

    Shun Nishide, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM     192 - 197  2009年  [査読有り]

     概要を見る

    Reliable predictability is one of the main factors that determine human behaviors. The authors developed a model that searches and generates robot motions based on reliable predictability. Training of the model consists of three phases. In the first phase, the model trains a sequential learner, namely Recurrent Neural Network with Parametric Bias, to self-organize robot and object dynamics. In the second phase, Steepest Descent Method is utilized to search for robot motion that induces the most predictable object motion. In the third phase, a hierarchical neural network is trained to link object image with the searched motion. Experiments were conducted with cylindrical objects. Analysis of the results have shown that the robot has acquired the most reliable robot motion, shifting it according to the posture of the object. Twenty motion generation experiments have resulted in generation of robot motion that induces consistent rolling motion of the objects. ©2009 IEEE.

    DOI

  • Development of a meeting browser towards supporting public involvement

    Shun Shiramatsu, Tadachika Ozono, Toramatsu Shintani, Kazunori Komatani, Tetsuya Ogata, Toru Takahashi, Hiroshi G. Okuno

    Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009   4   717 - 722  2009年

     概要を見る

    This paper presents novel methods for support for browsing a long meeting record towards supporting public involvement. Facilitating public involvement in the consensus building process for community development needs a lot of effort and time for sharing context and concerns among citizens and stakeholders. A record of public meeting often becomes too long to overview and to understand for people who did not participate in it. The two issues we addressed relate to how to best provide support for these people. First, support for overviewing the changes in a long meeting to track and to find intended arguments. Second, support for understanding the background of arguments. The approaches to the issues are first, to visualize the transition of topics in the meeting, and second provide information related to a transient topic specified by a user. The meeting browser we developed is designed on the basis of Visual Information-Seeking Mantra, "Overview first, zoom and filter, then details on demand." To visualize a dynamic topic flow, a graph for visualizing the topic transition, SalienceGraph is used to track the dynamic transition of the salience of a word. To visualize related information, the search engine based on SalienceGraph retrieves passages related to a transient topic from past meeting records or documents. These approaches support citizens and stakeholders to find, to track, and to understand a target argument from a long meeting record. © 2009 IEEE.

    DOI

  • Query-by-Example music retrieval approach based on musical genre shift by changing instrument volume

    Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 12th International Conference on Digital Audio Effects, DAFx 2009     205 - 212  2009年

     概要を見る

    We describe a novel Query-by-Example (QBE) approach in Music Information Retrieval, which allows a user to customize query examples by directly modifying the volume of different instrument parts. The underlying hypothesis is that the musical genre shifts (changes) in relation to the volume balance of different instruments. On the basis of this hypothesis, we aim to clarify the relationship between the change of the volume balance of a query and the shift in the musical genre of retrieved similar pieces, and thus help instruct a user in generating alternative queries without choosing other pieces. Our QBE system first separates all instrument parts from the audio signal of a piece with the help of its musical score, and then lets a user remix those parts to change acoustic features that represent musical mood of the piece. The distribution of those features is modeled by the Gaussian Mixture Model for each musical piece, and the Earth Movers Distance between mixtures of different pieces is used as the degree of their mood similarity. Experimental results showed that the shift was actually caused by the volume change of vocal, guitar, and drums.

  • Improving speech understanding accuracy with limited training data using multiple language models and multiple understanding models

    Masaki Katsumaru, Mikio Nakano, Kazunori Komatani, Kotaro Funakoshi, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH     2735 - 2738  2009年  [査読有り]

     概要を見る

    We aim to improve a speech understanding module with a small amount of training data. A speech understanding module uses a language model (LM) and a language understanding model (LUM). A lot of training data are needed to improve the models. Such data collection is, however, difficult in an actual process of development. We therefore design and develop a new framework that uses multiple LMs and LUMs to improve speech understanding accuracy under various amounts of training data. Even if the amount of available training data is small, each LM and each LUM can deal well with different types of utterances and more utterances are understood by using multiple LM and LUM. As one implementation of the framework, we develop a method for selecting the most appropriate speech understanding result from several candidates. The selection is based on probabilities of correctness calculated by logistic regressions. We evaluate our framework with various amounts of training data. Copyright © 2009 ISCA.

  • Thereminist Robot: Development of a Robot Theremin Player with Feedforward and Feedback Arm Control based on a Theremin”s Pitch Model (Invited paper)

    Takeshi MIZUMOTO, Hiroshi TSUJINO, Toru TAKAHASHI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS-2009)     2297 - 2302  2009年  [査読有り]

  • Changing timbre and phrase in existing musical performances as you like - Manipulations of single part using harmonic and inharmonic models

    Naoki Yasuraoka, Takehiro Abe, Katsutoshi Itoyama, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    MM'09 - Proceedings of the 2009 ACM Multimedia Conference, with Co-located Workshops and Symposiums     203 - 212  2009年

     概要を見る

    This paper presents a new music manipulation method that can change the timbre and phrases of an existing instrumental performance in a polyphonic sound mixture. This method consists of three primitive functions: 1) extracting and analyzing of a single instrumental part from polyphonic music signals, 2) mixing the instrument timbre with another, and 3) rendering a new phrase expression for another given score. The resulting customized part is re-mixed with the remaining parts of the original performance to generate new polyphonic music signals. A single instrumental part is extracted by using an integrated tone model that consists of harmonic and inharmonic tone models with the aid of the score of the single instrumental part. The extraction incorporates a residual model for the single instrumental part in order to avoid crosstalk between instrumental parts. The extracted model parameters are classified into their averages and deviations. The former is treated as instrument timbre and is customized by mixing, while the latter is treated as phrase expression and is customized by rendering. We evaluated our method in three experiments. The first experiment focused on introduction of the residual model, and it showed that the model parameters are estimated more accurately by 35.0 points. The second focused on timbral customization, and it showed that our method is more robust by 42.9 points in spectral distance compared with a conventional sound analysis-synthesis method, STRAIGHT. The third focused on the acoustic fidelity of customizing performance, and it showed that rendering phrase expression according to the note sequence leads to more accurate performance by 9.2 points in spectral distance in comparison with a rendering method that ignores the note sequence. Copyright 2009 ACM.

    DOI

  • Voice quality manipulation for humanoid robots consistent with their head movements

    Takuma Otsuka, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09     405 - 410  2009年

     概要を見る

    This paper presents voice-quality control of humanoid robots based on a new model of spectral envelope modification corresponding to the vertical head motions, and left-right sound-pressure modulation corresponding to the horizontal head motions. We assume that a pitch-axis rotation, or a vertical head motion, and a yaw-axis rotation, or a horizontal head motion, affect the voice quality independently. Spectral envelope modification model is constructed based on the analysis of human vocalizations. Left-right sound-pressure modulation model is established through the measurements of impulse responses using a pair of microphones. The experiments are carried out using two humanoid robots HRP-2 and Robovie-R2. Experimental results show that our method presents the change in the voice quality derived from pitch-axis head movement in a robot-to-robot dialogue situation when the interval between the robots are 50 cm. It is also confirmed that an observable modulation in the voice quality declines as the distance between the robots becomes large. The voice-cast directionality caused by yaw-axis rotation is observable using our model even when the robots stand as far as 150 cm away. ©2009 IEEE.

    DOI

  • Bowed string sequence estimation of a violin based on adaptive audio signal classification and context-dependent error correction

    Akira Maezawa, Katsutoshi Itoyama, Toru Takahashi, Tetsuya Ogata, Hiroshi G. Okuno

    ISM 2009 - 11th IEEE International Symposium on Multimedia     9 - 16  2009年  [査読有り]

     概要を見る

    The sequence of strings played on a bowed string instrument is essential to understanding of the fingering. Thus, its estimation is required for machine understanding of violin playing. Audio-based identification is the only viable way to realize this goal for existing music recordings. A naïve implementation using audio classification alone, however, is inaccurate and is not robust against variations in string or instruments. We develop a bowed string sequence estimation method by combining audio-based bowed string classification and context-dependent error correction. The robustness against different setups of instruments improves by normalizing the F0-dependent features using the average feature of a recording. The performance of error correction is evaluated using an electric violin with two different brands of strings and and an acoustic violin. By incorporating mean normalization, the recognition error of recognition accuracy due to changing the string alleviates by 8 points, and that due to change of instrument by 12 points. Error correction decreases the error due to change of string by 8 points and that due to different instrument by 9 points. © 2009 IEEE.

    DOI

  • ロボットによるRNNPBを用いた高予測信頼性動作探索とその解析

    西出俊, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第9回システムインテグレーション部門講演会(SI2008), 計測自動制御学会     81 - 82  2008年12月

  • 連続音響信号と構音情報の分節化に基づく母音音声模倣モデル

    神田尚, 尾形哲也, 高橋徹, 駒谷和範, 奥乃博

    第9回システムインテグレーション部門講演会(SI2008), 計測自動制御学会     639 - 640  2008年12月

  • RNNを用いた構音運動文節化に基づく連続母音模倣モデル

    神田尚, 尾形哲也, 高橋徹, 駒谷和範, 奥乃博

    (社)音響学会関西支部 第11回若手研究者交流研究発表会    2008年12月

  • 楽器音イコライザ:楽器パートの音量を操作可能なオーディオプレイヤー

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    (社)音響学会関西支部 第11回若手研究者交流研究発表会    2008年12月

  • Vocal Imitation Model with Segmenting and Composing Capability of Vowel Structure using Recurrent Neural Network

    Hisashi KANDA, Tetsuya OGATA, Toru TAKAHASHI, Kazunori KOMATANI, Hiroshi G. OKUNO

    第28回 AI チャレンジ研究会, SIG-Challenge-A802-2, 人工知能学会     7 - 12  2008年11月

  • Managing out-of-grammar utterances by topic estimation with domain extensibility in multi-domain spoken dialogue systems

    Kazunori Komatani, Satoshi Ikeda, Tetsuya Ogata, Hiroshi G. Okuno

    Speech Communication   50 ( 10 ) 863 - 870  2008年10月  [査読有り]

     概要を見る

    Spoken dialogue systems must inevitably deal with out-of-grammar utterances. We address this problem in multi-domain spoken dialogue systems, which deal with more tasks than a single-domain system. We defined a topic by augmenting a domain about which users want to find more information, and we developed a method of recovering out-of-grammar utterances based on topic estimation, i.e., by providing a help message in the estimated domain. Moreover, domain extensibility, that is, the ability to add new domains to the system, should be inherently retained in multi-domain systems. To estimate domains without sacrificing extensibility, we collected documents from the Web as training data. Since the data contained a certain amount of noise, we used latent semantic mapping (LSM), which enables robust topic estimation by removing the effects of noise from the data. Experimental results showed that our method improved topic estimation accuracy by 23.2 points for data including out-of-grammar utterances. © 2008 Elsevier B.V. All rights reserved.

    DOI

  • A game-theoretic model of referential coherence and its empirical verification using large Japanese and English corpora

    Shun Shiramatsu, Kazunori Komatani, Kôiti Hasida, Tetsuya Ogata, Hiroshi G. Okuno

    ACM Transactions on Speech and Language Processing   5 ( 3 ) 1 - 27  2008年10月

     概要を見る

    Referential coherence represents the smoothness of discourse resulting from topic continuity and pronominalization. Rational individuals prefer a referentially coherent structure of discourse when they select a language expression and its interpretation. This is a preference for cooperation in communication. By what principle do they share coherent expressions and interpretations? Centering theory is the standard theory of referential coherence [Grosz et al. 1995]. Although it is well designed on the bases of first-order inference rules [Joshi and Kuhn 1979], it does not embody a behavioral principle for the cooperation evident in communication. Hasida [1996] proposed a game-theoretic hypothesis in relation to this issue. We aim to empirically verify Hasida's hypothesis by using corpora of multiple languages. We statistically design language-dependent parameters by using a corpus of the target language. This statistical design enables us to objectively absorb language-specific differences and to verify the universality of Hasida's hypothesis by using corpora. We empirically verified our model by using large Japanese and English corpora. The result proves the language universality of the hypothesis.

    DOI

  • ミッシングフィーチャ理論に基づく複数話者同時発話音声認識における 音響特徴量とマスクの検討

    高橋徹, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    日本音響学会 2008年秋季研究発表会講演論文集     95 - 06  2008年10月

  • 独立成分分析に基づく適応フィルタのロボット聴覚への応用

    武田龍, 中臺一博, 駒谷和範, 尾形哲也

    日本ロボット学会誌   26 ( 6 ) 529 - 536  2008年09月

  • Synthesis Approach for Manipulating Pitch of a Musical Instrument Sound with Considering Timbral Characteristics

    Takehiro ABE, Katsutoshi ITOYAMA, Kazuyoshi YOSHII, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceeding of the 11th International Conference on Digital Audio Effects (DAFx-08)    2008年09月

  • Analysis of Reliable Predictability based Motion Generation using RNNPB

    Shun NISHIDE, Tetsuya OGATA, Jun TANI, Kazunori KOMATANI, Hiroshi G. OKUNO

    Proceedings of International Conference on Soft Computing and Intelligent Systems and International Symposium on advanced Intelligent Systems (SCIS&amp;ISIS 2008)     305 - 310  2008年09月

  • ロボット聴覚のためのソフトマスク生成法による周辺話者音声認識率の改善

    高橋徹, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    第26回日本ロボット学会学術講演会     1A1 - 01  2008年09月

  • 聴覚機能を持つ音楽ロボットのためのアーキテクチャの設計とビートカウントロボットへの適用

    水本武志, 武田龍, 吉井和佳, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    第26回日本ロボット学会学術講演会     1A1 - 02  2008年09月

  • ロボットにおける自己身体の順逆モデルを再利用した他者行為の予測と模倣

    尾形哲也, 横矢龍之介, 谷淳, 駒谷和範, 奥乃博

    第26回日本ロボット学会学術講演会     1J1 - 04  2008年09月

  • RNN を用いた連続音響信号からの母音構造と遷移情報の抽出

    神田尚, 尾形哲也, 駒谷和範, 奥乃博

    第26回日本ロボット学会学術講演会     1A2 - 01  2008年09月

  • 独立成分分析を応用したロボット聴覚による残響下におけるバージイン発話認識

    武田龍, 中臺一博, 高橋徹, 駒谷和範, 尾形哲也, 奥乃博

    第26回日本ロボット学会学術講演会     1A2 - 02  2008年09月

  • 物体挙動予測モデルによる動画像特徴量の自己組織化

    西出俊, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第26回日本ロボット学会学術講演会     2N2 - 04  2008年09月

  • 形態主張型コミュニケーションロボットにおける研究—形態主張行動の解り易さのユーザへの影響に関する研究—

    近藤裕樹, 奥出京司郎, 岩丸大二郎, 守良真, 菅佑樹, 尾形哲也, 菅野重樹

    第26回日本ロボット学会学術講演会     3J2 - 02  2008年09月

  • 音声対話システムにおけるラピッドプロトタイピングを指向したWFSTに基づく言語理解

    福林雄一朗, 駒谷和範, 中野幹生, 船越孝太郎, 辻野広司, 尾形哲也, 奥乃博

    情報処理学会論文誌   49 ( 8 ) 2762 - 2772  2008年08月

  • 音高による音色変化を考慮した楽器音の音高・音長操作手法

    安部武宏, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報科学研究会, 2008-MUS-76, 情報処理学会   Vol.2008  2008年08月

  • 楽器音イコライザによる音色の類似度に基づく楽曲検索システム

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報科学研究会, 2008-MUS-76, 情報処理学会   Vol.2008  2008年08月

  • 音素獲得に向けたリカレントニューラルネットワークによる音響信号と構音運動の分節化

    神田尚, 尾形哲也, 駒谷和範, 奥乃博

    日本機械学会ロボティクスメカトロニクス講演会     2P1 - G03  2008年06月

  • ユーザのカスタマイズを受容・拒否できる機構を持つロボットシステムの開発

    近藤裕樹, 坂上徳翁, 奥出京司郎, 岩丸大二郎, 菅佑樹, 尾形哲也, 菅野 重樹

    日本機械学会ロボティクスメカトロニクス講演会     1P1 - G11  2008年06月

  • 神経調節機能を参考とした自律エージェントの神経制御器の開発

    菅佑樹, 小林大三, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2P1 - G05  2008年06月

  • 物体操作に関する脳の情報処理構造を参考にした運動学習モデル

    有江浩明, 尾形哲也, 谷淳, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2P1  2008年06月

  • 自己組織化回路素子SONEの制御回路構造形成メカニズム

    金天海, 阿部博行, 出澤純一, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2P2 - G11  2008年06月

  • カスタマイズ可能なロボットにおける形態主張の効果と検証

    岩丸大二郎, 奥出京司郎, 近藤裕樹, 坂上徳翁, 菅佑樹, 尾形哲也, 菅野重樹

    人工知能学会全国大会 (JSAI2008)     1I1 - 04  2008年06月

  • : SalienceGraph: 参照確率に基づく話題遷移図の可視化

    白松俊, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会全国大会 (JSAI2008)     1H1 - 1  2008年06月

  • 音声対話システムにおける簡略表現認識のための誤認識増加を抑制する自動語彙拡張

    勝丸真樹, 駒谷和範, 尾形哲也, 奥乃 博

    第71回音声言語情報処理研究会, 情報処理学会     71 - 76  2008年05月

  • Predicting object dynamics from visual images through active sensing experiences

    Shun Nishide, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    Advanced Robotics   22 ( 5 ) 527 - 546  2008年04月  [査読有り]

     概要を見る

    Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the training phase, the authors use the recurrent neural network with parametric bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects and robot motor values are input into a hierarchical neural network to link the images to dynamic features (PB values). The neural network extracts prominent features that each induce object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated recursively. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. The results of the experiment predicting the dynamics of target objects proved that the technique is efficient for predicting the dynamics of the objects. © 2008 VSP.

    DOI

  • Motion from sound: Intermodal neural network mapping

    Tetsuya Ogata, Hiroshi G. Okuno, Hideki Kozima

    IEEE Intelligent Systems   23 ( 2 ) 76 - 78  2008年03月  [査読有り]

     概要を見る

    A technological method has been developed for intermodal mapping to generate robot motion from various sounds as well as to generate sounds from motions. The procedure consists of two phases, first the learning phase in which it observes some events together with associated sounds and then memorizes those sounds along with the motions of the sound source. Second phase is the interacting phase in which the robot receives limited sensory information from a single modality as input and associates this with different modality and expresses it. The recurrent-neural-network model with parametric bias (RNNPB) is applied that uses the current state-vector as input for outputting the next state-vector. The RNNPB model can self-organize the values that encode the input dynamics into special parametric-bias modes to reproduce he multimodal sensory flow.

    DOI

  • 楽譜情報を援用した多重奏音楽音響信号の音源分離と調波・非調波統合モデルの制約付パラメータ推定の同時実現

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   49 ( 3 ) 1465 - 1479  2008年03月

  • ロボットの順逆モデルの変換による他者行為予測と模倣

    横矢龍之介, 尾形哲也, 西出俊, 谷淳, 駒谷和範, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • ベース音高を考慮したポピュラー音楽に対する和音進行認識

    須見康平, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 楽器固有の音響的特徴を考慮した楽器音の音高操作手法

    安部武宏, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 複数楽器個体による事前分布を用いた調波・非調波統合モデルのパラメータ推定

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 音楽と自分の声を聞き分けながらビートに合わせて発声するロボットの開発

    水本武志, 武田龍, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 顔の動作に追従したGUIインタフェースを持つ音環境可視化システム

    久保田祐史, 吉田雅敏, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 音声対話システムにおけるユーザの固有名詞の簡略化に対処する語彙拡張

    勝丸真樹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • マルチドメインシステムにおけるトピック推定と対話履歴の統合によるドメイン選択の高精度化

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 音声対話システムにおけるWFSTに基づく文法検証を利用した動的ヘルプ生成

    福林雄一朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 音声対話システムにおける誤り原因の階層的分類とその推定に基づく発話誘導

    駒谷和範, 福林雄一朗, 池田智志, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • ロボット音声対話のためのMFTとICAによるバージイン許容機能の評価

    武田龍, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 音楽と映像の調和度計算モデルを用いたクロスメディア探索

    斎藤博己, 糸山克寿, 吉井和佳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 音源定位結果と音声認識結果をHMDに統合呈示する聴覚障害者向け音環境理解支援システム

    徳田浩一, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 神経回路モデルによる動作・言語変換を利用した人間ロボット音声協調

    張陽, 尾形哲也, 谷淳, 村瀬昌満, 駒谷和範, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • RNNPBによる音響模倣・分節化を用いた音素獲得モデルの提案

    神田尚, 尾形哲也, 駒谷和範, 奥乃博

    情報処理学会第70回全国大会    2008年03月

  • 新近性効果の減数曲線を加味した顕現性計算手法に基づく話題遷移の可視化

    白松俊, 駒谷和範, 尾形哲也, 奥乃博

    言語処理学会第14回年次大会     432 - 435  2008年03月

  • An efficient hybrid music recommender system using an incrementally trainable probabilistic generative model

    Kazuyoshi Yoshii, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE Transactions on Audio, Speech and Language Processing   16 ( 2 ) 435 - 447  2008年02月  [査読有り]

     概要を見る

    This paper presents a hybrid music recommender system that ranks musical pieces while efficiently maintaining collaborative and content-based data, i.e., rating scores given by users and acoustic features of audio signals. This hybrid approach overcomes the conventional tradeoff between recommendation accuracy and variety of recommended artists. Collaborative filtering, which is used on e-commerce sites, cannot recommend nonbrated pieces and provides a narrow variety of artists. Content-based filtering does not have satisfactory accuracy because it is based on the heuristics that the user's favorite pieces will have similar musical content despite there being exceptions. To attain a higher recommendation accuracy along with a wider variety of artists, we use a probabilistic generative model that unifies the collaborative and content-based data in a principled way. This model can explain the generative mechanism of the observed data in the probability theory. The probability distribution over users, pieces, and features is decomposed into three conditionally independent ones by introducing latent variables. This decomposition enables us to efficiently and incrementally adapt the model for increasing numbers of users and rating scores. We evaluated our system by using audio signals of commercial CDs and their corresponding rating scores obtained from an e-commerce site. The results revealed that our system accurately recommended pieces including nonrated ones from a wide variety of artists and maintained a high degree of accuracy even when new users and rating scores were added. © 2008 IEEE.

    DOI

  • 自己モデルの投影に基づくロボットによる他者発見と動作模倣

    横矢龍之介, 尾形哲也, 谷淳, 駒谷和範

    ヒューマンインタフェース学会論文誌   10 ( 1 ) 59 - 71  2008年02月

  • Hybrid Collaborative and Content-based Music Recommendation Using Incrementally-trainable Probabilistic Generative Model

    Kazuyoshi YOSHII, Masataka GOTO, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    IEEE Transactions on Audio, Speech and Language Processing   12 ( 2 ) 435 - 447  2008年02月

  • Meaning Games, In New Frontiers in Artificial Intelligence, JSAI 2007 Conference and Workshops

    Koiti HASHIDA, Shun SHIRAMATSU, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Revised Selected Papers, Lecture Notes in Artificial Intelligence   4914   228 - 241  2008年02月

  • Advanced Robotics: Preface

    Maria Chiara Carrozza, Tetsuya Ogata, Eugenio Guglielmelli

    Advanced Robotics   22 ( 1 ) 1 - 2  2008年01月

    DOI

  • Cheek to Chip: Dancing Robots and AI's Future.

    Jean-Julien Aucouturier, Katsushi Ikeuchi, Hirohisa Hirukawa, Shinichiro Nakaoka, Takaaki Shiratori, Shunsuke Kudoh, Fumio Kanehiro, Tetsuya Ogata, Hideki Kozima, Hiroshi G. Okuno, Marek P. Michalowski, Yuta Ogai, Takashi Ikegami, Kazuhiro Kosuge, Takahiro Takeda, Yasuhisa Hirata

    IEEE Intell. Syst.   23 ( 2 ) 74 - 84  2008年  [査読有り]

    DOI

  • Meaning games

    Kôiti Hasida, Shun Shiramatsu, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   4914 LNAI   228 - 241  2008年  [査読有り]

     概要を見る

    Communication can be accounted for in game-theoretic terms. The meaning game is proposed to formalize intentional communication in which the sender sends a message and the receiver attempts to infer its intended meaning. Using large Japanese and English corpora, the present paper demonstrates that centering theory is derived from a meaning game. This suggests that there are no language-specific rules on referential coherence. More generally speaking, language use seems to employ Pareto-optimal ESSs (evolutionarily stable strategies) of potentially very complex meaning games. There is still much to do before this complexity is elucidated in scientific terms, but game theory provides statistical and analytic means by which to advance the study on semantics and pragmatics of natural languages and other communication modalities. © 2008 Springer-Verlag Berlin Heidelberg.

    DOI

  • Structural feature extraction based on active sensing experiences

    Shun Nishide, Tetsuya Ogata, Ryunosuke Yokoya, Kazunori Komatani, Hiroshi G. Okuno, Jun Tani

    Proceedings - International Conference on Informatics Education and Research for Knowledge-Circulating Society, ICKS 2008     169 - 172  2008年  [査読有り]

     概要を見る

    Affordance is a feature of an object or environment that implies how to interact with it. Based on affordance theory, humans are said to perceive invariant structures for cognizing the object/environment for generating behaviors. In this paper, the authors present a method to extract invariant structures of objects from visual raw images, based on object manipulation experiences using a humanoid robot. The method consists of two training phases. The first phase utilizes Recurrent Neural Network with Parametric Bias (RNNPB) to self-organize dynamical object features extracted during active sensing with objects. The second phase trains a hierarchical neural network attached to RNNPB for associating object images and robot motions with self-organized object features. Analysis of the model has uncovered static objects features that are closely related to dynamic object motions, such as round or stable. © 2008 IEEE.

    DOI

  • Robot audition from the viewpoint of computational auditory scene analysis

    Hiroshi G. Okuno, Tetsuya Ogata, Kazunori Komatani

    Proceedings - International Conference on Informatics Education and Research for Knowledge-Circulating Society, ICKS 2008     35 - 40  2008年  [査読有り]

     概要を見る

    We have been engaged in research on computational auditory scene analysis to attain sophisticated robot/computer human interaction by manipulating real-world sound signals. The objective of our research is the understanding of an arbitrary sound mixture including music and environmental sounds as well as voiced speech, obtained by robot's ears (microphones) embedded on the robot. Three main issues in computational auditory scene analysis are sound source localization, separation, and recognition of separated sounds for a mixture of speech signals as well as polyphonic music signals. The Missing Feature Theory (MFT) approach integrates sound source separation and automatic speech recognition by generating missing feature masks. This robot audition system has been successfully ported to three kinds of robots, SIG2, Robovie R2 and Honda ASIMO. A robot recognizes three simultaneous speeches such as placing a meal order ora referee for RockPaper-Scissors Sound Games with a delay of less than 2 seconds. The real-time beat tracking system is also developed for robot audition. A robot hears music, understands and predicts its musical beats to behave in accordance with the beat times in real-time. © 2008 IEEE.

    DOI

  • Evaluation of two-channel-based sound source localization using 3D moving sound creation tool

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - International Conference on Informatics Education and Research for Knowledge-Circulating Society, ICKS 2008     209 - 212  2008年  [査読有り]

     概要を見る

    We proposed the way that can repeatedly evaluate the localization methods for moving sounds in the same condition regardless of a kind of methods and a number of microphones. And, we developed two-channel-based sound source localization integrated with a cross-power spectrum phase (CSP) analysis and EM algorithm. This one can localize several moving sounds and reduce localization error. Many sound source localization methods have already been developed. However, they could not be evaluated for moving sound in the same condition because it is hard to build database for moving sounds with accurate track information whenever making experiments. Also, to localize several moving sounds, conventional methods need a lot of microphone and/or prior information such as impulse response data. In this paper, we evaluated our sound localization method using 3D moving sound creation tool and confirmed that our method with two microphones can well localize the voices of a moving talker without impulse response data. © 2008 IEEE.

    DOI

  • Analysis-and-manipulation approach to pitch and duration of musical instrument sounds without distorting timbral characteristics

    Takehiro Abe, Katsutoshi Itoyama, Kazuyoshi Yoshii, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - 11th International Conference on Digital Audio Effects, DAFx 2008     249 - 256  2008年  [査読有り]

     概要を見る

    This paper presents an analysis-manipulation method that can generate musical instrument sounds with arbitrary pitches and durations from the sound of a given musical instrument (called seed) without distorting its timbrai characteristics. Based on psychoacoustical knowledge of the auditory effects of timbres, we defined timbrai features based on the spectrogram of the sound of a musical instrument as (i) the relative amplitudes of the harmonic peaks, (ii) the distribution of the inharmonic component, and (iii) temporal envelopes. First, to analyze the timbrai features of a seed, it was separated into harmonic and inharmonic components using Itoyama's integrated model. For pitch manipulation, we took into account the pitch-dependency of features (i) and (ii). We predicted the values of each feature by using a cubic polynomial that approximated the distribution of these features over pitches. To manipulate duration, we focused on preserving feature (iii) in the attack and decay duration of a seed. Therefore, only steady durations were expanded or shrunk. In addition, we propose a method for reproducing the properties of vibrato. Experimental results demonstrated the quality of the synthesized sounds produced using our method. The spectral and MFCC distances between the synthesized sounds and actual sounds of 32 instruments were reduced by 64.70% and 32.31%, respectively.

  • Vowel imitation using vocal tract model and recurrent neural network

    Hisashi Kanda, Tetsuya Ogata, Kazunori Komatani, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   4985 LNCS ( PART 2 ) 222 - 232  2008年  [査読有り]

     概要を見る

    A vocal imitation system was developed using a computational model that supports the motor theory of speech perception. A critical problem in vocal imitation is how to generate speech sounds produced by adults, whose vocal tracts have physical properties (i.e., articulatory motions) differing from those of infants' vocal tracts. To solve this problem, a model based on the motor theory of speech perception, was constructed. Applying this model enables the vocal imitation system to estimate articulatory motions for unexperienced speech sounds that have not actually been generated by the system. The system was implemented by using Recurrent Neural Network with Parametric Bias (RNNPB) and a physical vocal tract model, called Maeda model. Experimental results demonstrated that the system was sufficiently robust with respect to individual differences in speech sounds and could imitate unexperienced vowel sounds. © 2008 Springer-Verlag Berlin Heidelberg.

    DOI

  • Rapid Prototyping of Robust Language Understanding Modules for Spoken Dialogue Systems.

    Yuichiro Fukubayashi, Kazunori Komatani, Mikio Nakano, Kotaro Funakoshi, Hiroshi Tsujino, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Third International Joint Conference on Natural Language Processing (IJCNLP 2008)     210 - 216  2008年

  • Object dynamics prediction and motion generation based on reliable predictability

    Shun Nishide, Tetsuya Ogata, Ryunosuke Yokoya, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     1608 - 1614  2008年  [査読有り]

     概要を見る

    Consistency of object dynamics, which is related to reliable predictability, is an important factor for generating object manipulation motions. This paper proposes a technique to generate autonomous motions based on consistency of object dynamics. The technique resolves two issues: construction of an object dynamics prediction model and evaluation of consistency. The authors utilize Recurrent Neural Network with Parametric Bias to self-organize the dynamics, and link static images to the self-organized dynamics using a hierarchical neural network to deal with the first issue. For evaluation of consistency, the authors have set an evaluation function based on object dynamics relative to robot motor dynamics. Experiments have shown that the method is capable of predicting 90% of unknown object dynamics. Motion generation experiments have proved that the technique is capable of generating autonomous pushing motions that generate consistent rolling motions. ©2008 IEEE.

    DOI

  • Two-channel-based voice activity detection for humanoid robots in noisy home environments

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     3495 - 3501  2008年  [査読有り]

     概要を見る

    The purpose of this research is to accurately classify the speech signals originating from the front even in noisy home environments. This ability can help robots to improve speech recognition and to spot keywords. We therefore developed a new voice activity detection (VAD) based on the complex spectrum circle centroid (CSCC) method. It can classify the speech signals that are received at the front of two microphones by comparing the spectral energy of observed signals with that of target signals estimated by CSCC. Also, it can work in real time without training filter coefficients beforehand even in noisy environments (SNR > 0 dB) and can cope with speech noises generated by audio-visual equipments such as televisions and audio devices. Since the CSCC method requires the directions of the noise signals, we also developed a sound source localization system integrated with cross-power spectrum phase (CSP) analysis and an expectation-maximization (EM) algorithm. This system was demonstrated to enable a robot to cope with multiple sound sources using two microphones. ©2008 IEEE.

    DOI

  • A portable robot audition software system for multiple simultaneous speech signals

    H. G. Okuno, S. Yamamoto, K. Nakadai, J. M. Valin, T. Ogata, K. Komatani

    Proceedings - European Conference on Noise Control     483 - 488  2008年

     概要を見る

    Since a robot is deployed in various kinds of environments, the robot audition system should work with minimum prior information on environments to localize, separate and recognize utterances by multiple simultaneous talkers. For example, it should not assume either the number of speakers, the location of speakers for sound source separation (SSS), or specially tuned acoustic model for automatic speech recognition (ASR). We developed \HARK" portable robot audition that uses eight microphones installed on the surface of robot's body such as Honda ASIMO, and SIG-2 and Robovie-R2 at Kyoto University. HARK integrates SSS and ASR by using the Missing-Feature Theory. For SSS, we use Geometric Source Separation and multi-channel post-filter to separate each utterance. Since separated speech signals are distorted due to interfering talkers and sound source separation, multi-channel post-filter enhanced speech signals. At this process, we create a missing feature mask that specifies which acoustic features are reliable in time-frequency domain. Multi-band Julius, a missing-feature-theory based ASR, uses this mask to avoid the inuence of unreliable features in recognizing such distorted speech signals. The system demonstrated a waitress robot that accepts meal orders placed by three actual human talkers.

  • Integrating topic estimation and dialogue history for domain selection in multi-domain spoken dialogue systems

    Satoshi Ikeda, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   5027 LNAI   294 - 304  2008年  [査読有り]

     概要を見る

    We present a method of robust domain selection against out-of-grammar (OOG) utterances in multi-domain spoken dialogue systems. These utterances cause language-understanding errors because of a limited set of grammar and vocabulary of the systems, and deteriorate the domain selection. This is critical for multi-domain spoken dialogue systems to determine a system's response. We first define a topic as a domain from which the user wants to retrieve information, and estimate it as the user's intention. This topic estimation is enabled by using a large amount of sentences collected from the Web and Latent Semantic Mapping (LSM). The results are reliable even for OOG utterances. We then integrated both the topic estimation results and the dialogue history to construct a robust domain classifier against OOG utterances. The idea of integration is based on the fact that the reliability of the dialogue history is often impeded by language-understanding errors caused by OOG utterances, from which using topic estimation obtains useful information. Experimental results using 2191 utterances showed that our integrated method reduced domain selection errors by 14.3%. © 2008 Springer-Verlag Berlin Heidelberg.

    DOI

  • Development of user-adaptive value system of learning function using interactive EC

    Yuki Suga, Shigeki Sugano, Yoshinori Ikuma, Tetsuya Ogata

    IFAC Proceedings Volumes (IFAC-PapersOnline)   17 ( 1 PART 1 ) 9156 - 9161  2008年

     概要を見る

    Our goal is to create a user-adaptive communication-robot. We are developing a system for evaluating human-robot interactions. Although such evaluation is indispensable for learning algorithms, users' preferences are too difficult to model because they are subjective. In this study, we used the interactive evolutionary computation (IEC) to configure the value system of a learning communicationrobot. The IEC is a genetic algorithm whose fitness function is performed by the user. In our experiment, we encoded the values of sensors (reward or punishment) into genes, and subjects interacted with the learning robot. Through the interaction, the subjects evaluated the robot by touching its sensors, and the robot learned appropriate combinations between input and output. Afterward, the subjects gave their scores to the experimenter, and the scores were regarded as the fitness values of the corresponding genes. These sequences were continued until the 4 generation, and then the subjects compared three of their best genes and two of the experimenter's. We found that the user-adaptive value system is suitable for the communication-robot. Copyright © 2007 International Federation of Automatic Control All Rights Reserved.

    DOI

  • Human-Adaptive Robot Interaction Using Interactive EC with Human-Machine Hybrid Evaluation.

    Yuki Suga, Tetsuya Ogata, Shigeki Sugano

    J. Robotics Mechatronics   20 ( 4 ) 610 - 620  2008年

    DOI

  • Soft Missing-Feature Mask Generation for Simultaneous Speech Recognition System in Robots

    Toru Takahashi, Shun'ichi Yamamoto, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    INTERSPEECH 2008: 9TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2008, VOLS 1-5     992 - +  2008年  [査読有り]

  • Instrument equalizer for query-by-example retrieval: Improving sound source separation based on integrated harmonic and inharmonic models

    Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISMIR 2008 - 9th International Conference on Music Information Retrieval     133 - 138  2008年

     概要を見る

    This paper describes a music remixing interface, called Instrument Equalizer, that allows users to control the volume of each instrument part within existing audio recordings in real time. Although query-by-example retrieval systems need a user to prepare favorite examples (songs) in general, our interface gives a user to generate examples from existing ones by cutting or boosting some instrument/vocal parts, resulting in a variety of retrieved results. To change the volume, all instrument parts are separated from the input sound mixture using the corresponding standard MIDI file. For the separation, we used an integrated tone (timbre) model consisting of harmonic and inharmonic models that are initialized with template sounds recorded from a MIDI sound generator. The remaining but critical problem here is to deal with various performance styles and instrument bodies that are not given in the template sounds. To solve this problem, we train probabilistic distributions of timbre features by using various sounds. By adding a new constraint of maximizing the likelihood of timbre features extracted from each tone model, we succeeded in estimating model parameters that better express actual timbre.

  • Automatic chord recognition based on probabilistic integration of chord transition and bass pitch estimation

    Kouhei Sumi, Katsutoshi Itoyama, Kazuyoshi Yoshii, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISMIR 2008 - 9th International Conference on Music Information Retrieval     39 - 44  2008年

     概要を見る

    This paper presents a method that identifies musical chords in polyphonic musical signals. As musical chords mainly represent the harmony of music and are related to other musical elements such as melody and rhythm, the performance of chord recognition should improve if this interrelationship is taken into consideration. Nevertheless, this interrelationship has not been utilized in the literature as far as the authors are aware. In this paper, bass lines are utilized as clues for improving chord recognition because they can be regarded as an element of the melody. A probabilistic framework is devised to uniformly integrate bass lines extracted by using bass pitch estimation into a hypothesis-search-based chord recognition. To prune the hypothesis space of the search, the hypothesis reliability is defined as the weighted sum of three reliabilities: the likelihood of Gaussian Mixture Models for the observed features, the joint probability of chord and bass pitch, and the chord transition N-gram probability. Experimental results show that our method recognized the chord sequences of 150 songs in twelve Beatles albums; the average frame-rate accuracy of the results was 73.4%.

  • Segmenting acoustic signal with articulatory movement using recurrent neural network for phoneme acquisition

    Hisashi Kanda, Tetsuya Ogata, Kazunori Komatani, Hiroshi G. Okuno

    2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1712 - 1717  2008年  [査読有り]

     概要を見る

    This paper proposes a computational model for phoneme acquisition by infants. Human infants perceive speech sounds not as discrete phoneme sequences but as continuous acoustic signals. One of critical problems in phoneme acquisition is the design for segmenting these continuous speech sounds. The key idea to solve this problem is that articulatory mechanisms such as the vocal tract help human beings to perceive speech sound units corresponding to phonemes. That is, the ability to distinguish phonemes is learned by recognizing unstable points in the dynamics of continuous sound with articulatory movement. We have developed a vocal imitation system embodying the relationship between articulatory movements and sounds produced by the movements. To segment acoustic signal with articulatory movement, we apply the segmenting method to our system by Recurrent Neural Network with Parametric Bias (RNNPB). This method determines the multiple segmentation boundaries in a temporal sequence using the prediction error of the RNNPB model, and the PB values obtained by the method can be encoded as kind of phonemes. Our system was implemented by using a physical vocal tract model, called the Maeda model. Experimental results demonstrated that our system can self-organize the same phonemes in different continuous sounds. This suggests that our model reflects the process of phoneme acquisition. ©2008 IEEE.

    DOI

  • Target speech detection and separation for humanoid robots in sparse dialogue with noisy home environments

    Hyun Don Kim, Jinsung Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1705 - 1711  2008年  [査読有り]

     概要を見る

    In normal human communication, people face the speaker when listening and usually pay attention to the speaker' face. Therefore, in robot audition, the recognition of the front talker is critical for smooth interactions. This paper presents an enhanced speech detection method for a humanoid robot that can separate and recognize speech signals originating from the front even in noisy home environments. The robot audition system consists of a new type of voice activity detection (VAD) based on the complex spectrum circle centroid (CSCC) method and a maximum signal-to-noise (Max-SNR) beamformer. This VAD based on CSCC can classify speech signals that are retrieved at the frontal region of two microphones embedded on the robot. The system works in real-time without needing training filter coefficients given in advance even in a noisy environment (SNR > 0 dB). It can cope with speech noise generated from televisions and audio devices that does not originate from the center. Experiments using a humanoid robot, SIG2, with two microphones showed that our system enhanced extracted target speech signals more than 12 dB (SNR) and the success rate of automatic speech recognition for Japanese words was increased about 17 points. ©2008 IEEE.

    DOI

  • Design and evaluation of two-channel-based sound source localization over entire azimuth range for moving talkers

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     2197 - 2203  2008年

     概要を見る

    We propose a way to evaluate various sound localization systems for moving sounds under the same conditions. To construct a database for moving sounds, we developed a moving sound creation tool using the API library developed by the ARINIS Company. We developed a two-channel-based sound source localization system integrated with a cross-power spectrum phase (CSP) analysis and EM algorithm. The CSP of sound signals obtained with only two microphones is used to localize the sound source without having to use prior information such as impulse response data. The EM algorithm helps the system cope with several moving sound sources and reduce localization error. We evaluated our sound localization method using artificial moving sounds and confirmed that it can well localize moving sounds slower than 1.125 rad/sec. Finally, we solve the problem of distinguishing whether sounds are coming from the front or back by rotating a robot's head equipped with only two microphones. Our system was applied to a humanoid robot called SIG2, and we confirmed its ability to localize sounds over the entire azimuth range. ©2008 IEEE.

    DOI

  • A robot listens to music and counts its beats aloud by separating music from counting voice

    Takeshi Mizumoto, Ryu Takeda, Kazuyoshi Yoshii, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1538 - 1543  2008年

     概要を見る

    This paper presents a beat-counting robot that can count musical beats aloud, i.e., speak "one, two, three, four, one, two, ..." along music, while listening to music by using its own ears. Music-understanding robots that interact with humans should be able not only to recognize music internally, but also to express their own internal states. To develop our beat-counting robot, we have tackled three issues: (1) recognition of hierarchical beat structures, (2) expression of these structures by counting beats, and (3) suppression of counting voice (self-generated sound) in sound mixtures recorded by ears. The main issue is (3) because the interference of counting voice in music causes the decrease of the beat recognition accuracy. So we designed the architecture for music-understanding robot that is capable of dealing with the issue of self-generated sounds. To solve these issues, we took the following approaches: (1) beat structure prediction based on musical knowledge on chords and drums, (2) speed control of counting voice according to music tempo via a vocoder called STRAIGHT, and (3) semi-blind separation of sound mixtures into music and counting voice via an adaptive filter based on ICA (Independent Component Analysis) that uses the waveform of the counting voice as a prior knowledge. Experimental result showed that suppressing robot's own voice improved music recognition capability. ©2008 IEEE.

    DOI

  • Barge-in-able robot audition based on ICA and missing feature theory under semi-blind situation

    Ryu Takeda, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1718 - 1723  2008年  [査読有り]

     概要を見る

    This paper describes a robot audition system that allows the user to barge-in; that is, the user can speak simultaneously when the robot is speaking. Our "barge-in-able" system consists of two stages: (1) cancellation of robot speech and (2) recognition of the separated user speech under the "semi-blind situation". The semi-blind situation is where a robot's speech signal is known but a user's speech signal is not. The first stage is achieved by using an adaptive filter based on time-frequency domain Independent Component Analysis, because that can separate robot speech more robustly against noise than conventional echo cancellers. To improve performance in online processing, we utilized known source normalization and the exponentially weighted stepsize method. The second stage is achieved by automatic speech recognition (ASR) based on the missing feature theory which provides robust recognition by exploiting the reliability of speech features distorted due to noise and/or separation. The semi-blind situation simplifies the estimation of such reliabilities. Experiments demonstrated that our system improved word correctness of ASR by 10.0 %. ©2008 IEEE.

    DOI

  • Active sensing based dynamical object feature extraction

    Shun Nishide, Tetsuya Ogata, Ryunosuke Yokoya, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    2008 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1 - 7  2008年  [査読有り]

     概要を見る

    This paper presents a method to autonomously extract object features that describe their dynamics from active sensing experiences. The model is composed of a dynamics learning module and a feature extraction module. Recurrent Neural Network with Parametric Bias (RNNPB) is utilized for the dynamics learning module, learning and self-organizing the sequences of robot and object motions. A hierarchical neural network is linked to the input of RNNPB as the feature extraction module for extracting object features that describe the object motions. The two modules are simultaneously trained using image and motion sequences acquired from the robot's active sensing with objects. Experiments are performed with the robot's pushing motion with a variety of objects to generate sliding, falling over, bouncing, and rolling motions. The results have shown that the model is capable of extracting features that distinguish the characteristics of object dynamics. ©2008 IEEE.

    DOI

  • Extensibility verification of robust domain selection against out-of-grammar utterances in multi-domain spoken dialogue system

    Satoshi Ikeda, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH     487 - 490  2008年  [査読有り]

     概要を見る

    We developed a robust domain selection method and verified its extensibility. An issue in domain selection is its robustness against out-of-grammar utterances. It is essential to generate correct system responses because such utterances often cause domain selection errors. We therefore integrated the topic estimation results and the dialogue history to construct a robust domain classifier. Another issue is that domain selection should be performed within an extensible framework, because the system is often modified and extended. That is, the classifier should still have high performance without reconstructing it after adding new domains. The extensibility of our method was not experimentally verified yet, because it requires a lot of effort to collect new dialogue data after extending the system. Therefore, we verified extensibility without collecting new data. We constructed the classifier by leaving out some domains in the dialogue data and then evaluated its accuracy as the classifier for the data where the left-out domains were virtually added. Copyright © 2008 ISCA.

  • Expanding Vocabulary for Recognizing User”s Abbreviations of Proper Nouns without Increasing ASR Error Rates in Spoken Dialogue Systems

    Masaki KATSUMARU, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of International Conference on Spoken Language Processing (Interspeech-2008)     187 - 190  2008年  [査読有り]

  • Reinforcement Signal Propagation Algorithm for Logic Circuit.

    Chyon Hae Kim, Tetsuya Ogata, Shigeki Sugano

    J. Robotics Mechatronics   20 ( 5 ) 757 - 774  2008年

    DOI

  • SalienceGraph: Visualizing salience dynamics of written discourse by using reference probability and PLSA

    Shun Shiramatsu, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   5351 LNAI   890 - 902  2008年  [査読有り]

     概要を見る

    Since public involvement in the decision-making process for community development needs a lot of efforts and time, support tools for speeding up the consensus building process among stakeholders are required. This paper presents a new method for finding, tracking and visualizing participants' concerns (topics) from the record of a public debate. For finding topics, we use the salience of a term, which is computed as its reference probability based on referential coherence in the Centering Theory. Our system first annotates a debate record or minute into Global Document Annotation (GDA) format automatically, and then computes the salience of each term from the GDA-annotated text sentence by sentence. Then, by using the Probalilistic Latent Semantic Analytsis (PLSA), our system reduces the dimensions of the vector of salience values of terms into a set of major latent topics. For tracking topics, we use the salience dynamics, which is computed as the temporal change of joint attention to the major latent topics with additional user-supplied terms. The resulting graph is called SalienceGraph. For visualizing SalienceGraph, we use 3D visualizer with GUI designed by "overview first, zoom and filter, then details on demand" principle. SalienceGraph provides more accurate trajectory of topics than conventional TF•IDF. © 2008 Springer Berlin Heidelberg.

    DOI

  • Design and implementation of 3d auditory scene visualizer towards auditory awarenesswith face tracking

    Yuji Kubota, Masatoshi Yoshida, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - 10th IEEE International Symposium on Multimedia, ISM 2008     468 - 476  2008年  [査読有り]

     概要を見る

    If machine audition can recognize an auditory scene containing simultaneous and moving talkers, what kinds of awareness will people gain from an auditory scene visualizer? This paper presents the design and implementation of 3D Auditory Scene Visualizer based on the visual information seeking mantra, i.e., "overview first, zoom and filter, then details on demand". The machine audition system called HARK captures 3D sounds with a microphone array, localizes and separates sounds, and recognizes separated sounds by automatic speech recognition (ASR). The 3D visualizer implemented in Java 3D displays each sound stream as a beam originating from the center of the microphones (overview mode), shows temporal snapshots with/without specifying focusing areas (zoom and filter mode), and shows detailed information about a particular sound stream (details on demand). In the details-ondemand mode, ASR results are displayed in a "karaoke" manner, i.e., character-by-character. This three-mode visualization will give the user auditory awareness enhanced by HARK. In addition, a face-tracking system automatically changes the focus of attention by tracking the user's face. The resulting system is portable and can be deployed in any place, so it is expected to give more vivid awareness than expensive high-fidelity auditory scene reproduction systems. © 2008 IEEE.

    DOI

  • 3D Auditory Scene Visualizer with face tracking: Design and implementation for auditory awareness compensation

    Yuji Kubota, Shun Shiramatsu, Masatoshi Yoshida, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 2nd International Symposium on Universal Communication, ISUC 2008     42 - 49  2008年

     概要を見る

    This paper presents the design and implementation of 3D Auditory Scene Visualizer based on the visual information seeking mantra, "overview first, zoom and filter, then details on demand". The machine audition system called HARK captures 3D sounds with a microphone array. The natural language processing called SalienceGraph visualizes topic transition by using discourse salience. The 3D visualizer implemented in Java 3D displays topic transition and each sound stream as a beam originating from the microphones (overview mode), shows temporal snapshots with/without specifying focusing areas (zoom-andfilter mode), and shows detailed information about a particular sound stream (details-on-demand mode). This threemode visualization will give the user auditory awareness enhanced by HARK and SalienceGraph. In addition, a facetracking system automatically determines the user's intention by tracking the user's face. The resulting system will enable users to manage and browse auditory scene files effectively, so it should acceleration and support the information explosion to compensate the lack of auditory awareness. © 2008 IEEE.

    DOI

  • Soft missing-feature mask generation for simultaneous speech recognition system in robots

    Toru Takahashi, Shun'ichi Yamamoto, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH   1 ( 1 ) 992 - 995  2008年  [査読有り]

     概要を見る

    This paper addresses automatic soft missing-feature mask (MFM) generation based on a leak energy estimation for a simultaneous speech recognition system. An MFM is used as a weight for probability calculation in a recognition process. In a previous work, a threshold-base-zero-or-one function was applied to decide if spectral parameter can be reliable or not for each frequency bin. The function is extended into a weighted sigmoid function which has two free parameters. In addition, a contribution ratio of static features is introduced for the probability calculation in a recognition process which static and dynamic features are input. The ratio can be implemented as a part of soft mask. The average recognition rate based on a soft MFM improved by about 5% for all directions from a conventional system based on a hard MFM. Word recognition rates improved from 70 to 80% for peripheral talkers and from 93 to 97% for front speech when speakers were 90 degrees apart. Copyright © 2008 ISCA.

    DOI

  • Experience-based imitation using RNNPB

    Ryunosuke Yokoya, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    Advanced Robotics   21 ( 12 ) 1351 - 1367  2007年12月  [査読有り]

     概要を見る

    Robot imitation is a useful and promising alternative to robot programming. Robot imitation involves two crucial issues. The first is how a robot can imitate a human whose physical structure and properties differ greatly from its own. The second is how the robot can generate various motions from finite programmable patterns (generalization). This paper describes a novel approach to robot imitation based on its own physical experiences. We considered the target task of moving an object on a table. For imitation, we focused on an active sensing process in which the robot acquires the relation between the object's motion and its own arm motion. For generalization, we applied the RNNPB (recurrent neural network with parametric bias) model to enable recognition/generation of imitation motions. The robot associates the arm motion which reproduces the observed object's motion presented by a human operator. Experimental results proved the generalization capability of our method, which enables the robot to imitate not only motion it has experienced, but also unknown motion through nonlinear combination of the experienced motions. © 2007 VSP.

    DOI

  • 能動知覚経験に基づく物体挙動連想と動作生成

    西出俊, 尾形哲也, 横矢龍之介, 谷淳, 駒谷和範, 奥乃博

    第8回システムインテグレーション部門講演会 (SI2007), 計測自動制御学会     1C1 - 3  2007年12月

  • ニューラルネットによる腱駆動ロボットアームの制御

    有江浩明, 尾形哲也, 谷淳, 菅野重樹

    第8回システムインテグレーション部門講演会 (SI2007) ,計測自動制御学会     1B4 - 6  2007年12月

  • 自己モデルの再利用に基づくロボットによる他者の発見と模倣

    横矢龍之介, 尾形哲也, 西出俊, 谷淳, 駒谷和範, 奥乃博

    第8回システムインテグレーション部門講演会 (SI2007), 計測自動制御学会     2C1 - 2  2007年12月

  • マルチドメインシステムにおけるトピック推定と対話履歴の統合によるドメイン選択の高精度化

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    音声言語シンポジウム, 信学技法   NLC2007-80 ( SP2007-143 ) 277 - 282  2007年12月

  • 実世界の力学構造に基づいた疑似シンボル生成と言語動作相互変換

    尾形哲也, 谷淳, 駒谷和範, 奥乃博

    システム・情報部門学術講演会講演論文集 (SSI2007), 計測自動制御学会   2A2-3   211 - 216  2007年11月

  • Advanced Robotics: Preface

    Maria Chiara Carrozza, Tetsuya Ogata, Eugenio Guglielmelli

    Advanced Robotics   21 ( 10 ) 1093 - 1095  2007年10月

    DOI

  • Reinforcement learning of a continuous motor sequence with hidden states

    Hiroaki Arie, Tetsuya Ogata, Jun Tani, Shigeki Sugano

    Advanced Robotics   21 ( 10 ) 1215 - 1229  2007年10月  [査読有り]

     概要を見る

    Reinforcement learning is the scheme for unsupervised learning in which robots are expected to acquire behavior skills through self-explorations based on reward signals. There are some difficulties, however, in applying conventional reinforcement learning algorithms to motion control tasks of a robot because most algorithms are concerned with discrete state space and based on the assumption of complete observability of the state. Real-world environments often have partial observablility; therefore, robots have to estimate the unobservable hidden states. This paper proposes a method to solve these two problems by combining the reinforcement learning algorithm and a learning algorithm for a continuous time recurrent neural network (CTRNN). The CTRNN can learn spatio-temporal structures in a continuous time and space domain, and can preserve the contextual flow by a self-organizing appropriate internal memory structure. This enables the robot to deal with the hidden state problem. We carried out an experiment on the pendulum swing-up task without rotational speed information. As a result, this task is accomplished in several hundred trials using the proposed algorithm. In addition, it is shown that the information about the rotational speed of the pendulum, which is considered as a hidden state, is estimated and encoded on the activation of a context neuron. © 2007 VSP.

    DOI

  • ゲーム理論に基づく参照結束性のモデル化と日本語・英語の大規模コーパスを用いた統計的検証

    白松俊, 駒谷和範, 橋田浩一, 尾形哲也, 奥乃博

    自然言語処理   14 ( 5 ) 199 - 239  2007年10月

  • 自己組織化回路素子SONEにおけるノイズの抑制

    金天海, 出澤純一, 尾形哲也, 菅野重樹

    日本ロボット学会誌   25 ( 6 ) 115 - 122  2007年09月

  • Hybrid Collaborative and Content-based Music Recommendation Using Probabilistic Model with Latent User Preferences

    Kazuyoshi YOSHII, Masataka GOTO, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of 8th International Conference on Musical Information Retreival (ISMIR-2007)     89 - 94  2007年09月

  • 独立成分分析に基づく適応フィルタのロボット聴覚への応用

    武田龍, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    第25回日本ロボット学会学術講演会     1N16  2007年09月

  • 声道物理モデルとリカレントニューラルネットワークによる母音模倣

    神田尚, 尾形哲也, 駒谷和範, 奥乃博

    第25回日本ロボット学会学術講演会     1N17  2007年09月

  • リカレントニューラルネットワークによる複数文章とロボット動作の双方向変換

    尾形哲也, 村瀬昌満, 谷淳, 駒谷和範, 奥乃博

    第25回日本ロボット学会学術講演会     1C26  2007年09月

  • 物体挙動の予測信頼性に基づく自律的な動作生成

    西出俊, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第25回日本ロボット学会学術講演会     1C36  2007年09月

  • 対話型進化的計算による強化学習器の報酬系の獲得

    菅佑樹, 生熊良規, 尾形哲也, 菅野重樹

    第25回日本ロボット学会学術講演会,2O18,    2007年09月

  • 自己モデルの投影に基づくロボットによる模倣動作の自律的獲得

    横矢龍之介, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第25回日本ロボット学会学術講演会     3N27  2007年09月

  • マルチドメイン音声対話システムにおける想定外発話への対処のためのWebを用いたシステム知識の拡張

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    情報技術レターズ (第6回情報科学技術フォーラム(FIT2007)講演論文集)     LE - 007  2007年09月

  • 音色特徴量分布の利用による調波・非調波併用モデルのパラメータ推定

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会 音楽情報処理研究会 研究報告, 2007-SIG71-26    2007年08月

  • ドメイン拡張性を備えたトピック推定に基づく発話誘導を行うマルチドメイン音声対話システム

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会研究会資料 SIG-SLUD-A701-10 (7/23)     83 - 88  2007年07月

  • 会話文脈に応じた関連情報提示タスクのための文脈類似度計算手法の開発

    白松俊, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会研究会資料 SIG-SLUD-A701-10 (7/23)     57 - 62  2007年07月

  • 自己組織化回路素子へのフリップフロップ素子導入による時系列学習

    金天海, 出澤純一, 尾形哲也, 菅野重樹

    第21回人工知能学会全国大会     3G6 - 1  2007年06月

  • マルチドメイン音声対話システムにおける対話履歴を利用したドメイン選択

    神田直之, 駒谷和範, 中野幹生, 中臺一博, 辻野広司, 尾形哲也, 奥乃博

    情報処理学会論文誌   48 ( 5 ) 1980 - 1989  2007年05月

  • Meaning-Game-based Centering Model with Statistical Definition of Utility of Referential Expression and Its Verification Using Japanese and English Corpora

    Shun SHIRAMATSU, Kazunori KOMATANI, Koiti HASIDA, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of the 6th Discourse Anaphora and Anaphor Resolution Colloquium (DAARC2007)     121 - 126  2007年05月

  • Drumix: An Audio Player with Functions of Realtime Drum-Part Rearrangement for Active Music Listening

    Kazuyoshi YOSHII, Masataka GOTO, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Journal of Information Processing Society of Japan   48 ( 3 ) 1229 - 1239  2007年03月

  • ICA と MFT に基づく音声認識における Soft Mask を用いた性能評価

    武田龍, 山本俊一, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     6ZB - 6  2007年03月

  • 音環境を可視化する録音再生システム

    吉田雅敏, 海尻聡, 山本俊一, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     6ZB - 2  2007年03月

  • 物体静止画像から動的特徴を抽出する神経回路モデルの学習と解析

    西出俊, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    情報処理学会第69回全国大会     6B - 4  2007年03月

  • マルチドメイン音声対話システムにおけるシステム想定外発話のトピック推定に基づく発話誘導

    池田智志, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     5Q - 6  2007年03月

  • 音声対話システムにおける発話検証を利用したシステム想定外発話の誤受理抑制

    福林雄一朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     5Q - 5  2007年03月

  • EMアルゴリズムとパーティクルフィルタの統合によるリアルタイム複数の人物追跡システム実現

    金鉉燉, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     4B - 3  2007年03月

  • 聴覚障害児の授業支援のためのHMDによる音声認識結果提示システムの設計

    徳田浩一, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     3ZB - 4  2007年03月

  • Onomatree:擬音語と木構造を併用した環境音検索インターフェース

    清水敬太, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     3N - 7  2007年03月

  • 自己組織化回路素子(SONE)への教師あり学習の付与

    金天海, 尾形哲也, 菅野重樹

    情報処理学会第69回全国大会     2Q - 3  2007年03月

  • マルチメディアコンテンツにおける音楽と映像の調和に関する分析

    西山正紘, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     2N - 6  2007年03月

  • 楽譜情報を用いたNMFによる音楽音響信号の音源分離

    糸山克寿, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第69回全国大会     2N - 1  2007年03月

  • RNNPBによる自然言語列と動作列の意味的結合と人間ロボットインタラクション

    村瀬昌満, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    情報処理学会第69回全国大会     1R - 4  2007年03月

  • 自己身体モデルの投影に基づくロボットによる他者の動作予測

    横矢龍之介, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    情報処理学会第69回全国大会     1R - 3  2007年03月

  • 人間型声道モデルと神経回路モデルを利用した母音模倣

    神田尚, 尾形哲也, 駒谷和範, 奥乃博

    情報処理学会第69回全国大会     1Q - 2  2007年03月

  • 可聴域音波を用いたAcoustical Holography (AH) による遮蔽物の検出と距離計測法

    丹羽治彦, 駒谷和範, 尾形哲也, 奥乃博

    音響学会春季講演会     1  2007年03月

  • 歌声の分離と音響モデルの分離歌声への適用に基づく音楽音響信号と歌詞の時間的対応付け手法

    藤原弘将, 後藤真孝, 緒方淳, 駒谷和範, 尾形哲也, 奥乃博

    音響学会春季講演会     3  2007年03月

  • コーパスからの関連語獲得に基づく連想を加味した顕現性の推定

    白松俊, 駒谷和範, 尾形哲也, 奥乃博

    言語処理学会第13回年次大会    2007年03月

  • 音声認識結果とコンセプトへの重みづけによるWFSTに基づく音声言語理解の高精度化

    福林雄一朗, 駒谷和範, 中野幹生, 船越孝太郎, 辻野広司, 尾形哲也, 奥乃博

    第66回音声言語情報処理研究会, 2007-SLP-66 (8), 2007-NL-179 (8) 情報処理学会   2007 ( 47 ) 43 - 48  2007年03月

  • 多重奏音楽音響信号の音源分離のための調波・非調波モデルの制約付きパラメータ推定

    糸山克寿, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報処理研究会, 2007-MUS-70 (13), 2007-EC-7 (13)   2007 ( 37 ) 81 - 88  2007年03月

  • マルチメディアコンテンツにおける音楽と映像の調和度計算モデル

    西山正紘, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    音楽情報処理研究会, 2007-MUS-,Vol.2007, No., 情報処理学会    2007年02月

  • 音声対話システムにおけるヘルプ生成のためのシステム想定外発話の誤受理抑制

    福林雄一朗, 駒谷和範, 尾形哲也, 奥乃博

    第65回音声言語情報処理研究会, Vol.2007, No., 情報処理学会    2007年02月

  • Computational auditory scene analysis and its application to robot audition: Five years experience

    Hiroshi G. Okuno, Tetsuya Ogata, Kazunori Komatani

    Proceedings - Second International Conference on Informatics Research for Development of Knowledge Society Infrastructure, ICKS 2007     69 - 76  2007年  [査読有り]

     概要を見る

    We have been engaged in research on computational auditory scene analysis to attain sophisticated robot/computer human interaction by manipulating real-world sound signals. The objective of our research is the understanding of an arbitrary sound mixture including non-speech sounds and music as well as voiced speech, obtained by robot's ears, that is, microphones embedded in the robot. We have coped with three main issues in computational auditory scene analysis, that is, sound source localization, separation, and recognition of separated sounds for a mixture of speech signals as well as polyphonic music signals. This paper overviews our results in robot audition, in particular, Missing Feature Theory based integration of sound source separation and automatic speech recognition, and those in music information processing, in particular, drum sound equalizer. © 2007 IEEE.

    DOI

  • Instrogram: Probabilistic Representation of Instrument Existence for Polyphonic Music

    Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ipsjdc   3   1  2007年  [査読有り]

    DOI

  • Improving efficiency and scalability of model-based music recommender system based on incremental training

    Kazuyoshi Yoshii, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007     89 - 94  2007年  [査読有り]

     概要を見る

    We aimed at improving the efficiency and scalability of a hybrid music recommender system based on a probabilistic generative model that integrates both collaborative data (rating scores provided by users) and content-based data (acoustic features of musical pieces). Although the hybrid system was proved to make accurate recommendations, it lacks efficiency and scalability. In other words, the entire model needs to be re-trained from scratch whenever a new score, user, or piece is added. Furthermore, the system cannot deal with practical numbers of users and pieces on an enterprise scale. To improve efficiency, we propose an incremental method that partially updates the model at low computational cost. To enhance scalability, we propose a method that first constructs a small "core" model over fewer virtual representatives created from real users and pieces, and then adds the real users and pieces to the core model by using the incremental method. The experimental results revealed that the proposed system was not only efficient and scalable but also outperformed the original system in terms of accuracy. ©2007 Austrian Computer Society (OCG).

  • Drumix: An Audio Player with Real-time Drum-part Rearrangement Functions for Active Music Listening

    Kazuyoshi Yoshii, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ipsjdc   3   134  2007年  [査読有り]

    DOI

  • Instrument identification in polyphonic music: Feature weighting to minimize influence of sound overlaps

    Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Eurasip Journal on Advances in Signal Processing   2007  2007年  [査読有り]

     概要を見る

    We provide a new solution to the problem of feature variations caused by the overlapping of sounds in instrument identification in polyphonic music. When multiple instruments simultaneously play, partials (harmonic components) of their sounds overlap and interfere, which makes the acoustic features different from those of monophonic sounds. To cope with this, we weight features based on how much they are affected by overlapping. First, we quantitatively evaluate the influence of overlapping on each feature as the ratio of the within-class variance to the between-class variance in the distribution of training data obtained from polyphonic sounds. Then, we generate feature axes using a weighted mixture that minimizes the influence via linear discriminant analysis. In addition, we improve instrument identification using musical context. Experimental results showed that the recognition rates using both feature weighting and musical context were 84.1 % for duo, 77.6 % for trio, and 72.3 % for quartet; those without using either were 53.4, 49.6, and 46.5 % , respectively.

    DOI

  • 音源分離との統合によるミッシングフィーチャマスク自動生成に基づく同時発話音声認識

    山本俊一, 中臺一博, 中野幹生, 辻野広司, Jean-Marc Valin, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会誌   25 ( 1 ) 92 - 102  2007年01月

  • Instrogram: Probabilistic Representation of Instrument Existence for Polyphonic Music

    Tetsuro KITAHARA, Masataka GOTO, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Journal of Information Processing Society of Japan   48 ( 1 ) 214 - 216  2007年01月

  • Integration and adaptation of harmonic and inharmonic models for separating polyphonic musical signals

    Katsutoshi Itoyama, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings   1   57 - 60  2007年  [査読有り]

     概要を見る

    This paper describes a sound source separation method for polyphonic sound mixtures o music to build an instrument equalizer for remixing multiple tracks separated from compact-disc recordings by changing the volume level of each track. Although such, mixtures usually include both harmonic and inharmonic sounds, the difficulties in dealing with both types of sounds together have not been addressed in most previous methods that have focused on either of the two types separately. We therefore developed an integrated weighted-mixture model consisting of both harmonic-structure and inharmonic-structure tone models (generative models for the power spectrogram). On the basis of the MAP estimation using the EM algorithm, we estimated all model parameters of this, integrated model under several original constraints for preventing over-training and maintaining intra-instrument consistency. Using standard MIDI files as prior information of the model parameters, We applied this model to compact-disc recordings and achieved the instrument equalizer. © 2007 IEEE.

    DOI

  • Enhancement of self organizing network elements for supervised learning

    Chyon Hae Kim, Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Conference on Robotics and Automation     92 - 98  2007年  [査読有り]

     概要を見る

    We have proposed self-organizing network elements (SONE) as a learning method for robots to meet the requirements of autonomous exploration of effective output, simple external parameters, and low calculation costs. SONE can be used as an algorithm for obtaining network topology by propagating reinforcement signals between the elements of a network. Traditionally, the analysis of fundamental features in SONE and their application to supervised learning tasks were difficult because the learning method of SONE was limited to reinforcement learning. Here the abilities of generalization, incremental learning, and temporal sequence learning were evaluated using a supervised learning method with SONE. Moreover, the proposed method enabled our SONE to be applied to a greater variety of tasks. © 2007 IEEE.

    DOI

  • Distance estimation of hidden objects based on acoustical holography by applying acoustic diffraction of audible sound

    Haruhiko Niwa, Tetsuya Ogata, Kazunori Komatani, Okuno G. Hiroshi

    Proceedings - IEEE International Conference on Robotics and Automation     423 - 428  2007年  [査読有り]

     概要を見る

    Occlusion is a problem for range finders; ranging systems using cameras or lasers cannot be used to estimate distance to an object (hidden object) that is occluded by another (obstacle). We developed a method to estimate the distance to the hidden object by applying acoustic diffraction of audible sound. Our method is based on time-of-flight (TOF), which has been used in ultrasound ranging systems. We determined the best frequency of audible sound and designed its optimal modulated signal for our system. We determined that the system estimates the distance to the hidden object as well as the obstacle. However, the measurement signal obtained from the hidden object was weak. Thus, interference from sound signals reflected from other objects or walls was not negligible. Therefore, we combined acoustical holography (AH) and TOF, which enabled a partial analysis of the reflection sound intensity field around the obstacle and hidden object. Our method was effective for ranging two objects of the same size within a 1.2 m depth range. The accuracy of our method was 3 cm for the obstacle, and 6 cm for the hidden object. ©2007 IEEE.

    DOI

  • Human-robot cooperation using quasi-symbols generated by RNNPB model

    Tetsuya Ogata, Shohei Matsumoto, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     2156 - 2161  2007年  [査読有り]

     概要を見る

    We describe a means of human robot interaction based not on natural language but on "quasi symbols," which represent sensory-motor dynamics in the task and/or environment. It thus overcomes a key problem of using natural language for human-robot interaction - the need to understand the dynamic context The quasi-symbols used are motion primitives corresponding to the attractor dynamics of the sensory-motor flow. These primitives are extracted from the observed data using the recurrent neural network with parametric bias (RNNPB) model. Binary representations based on the model parameters were implemented as quasi symbols in a humanoid robot, Robovie. The experiment task was robot-arm operation on a table. The quasi-symbols acquired by learning enabled the robot to perform novel motions. A person was able to control the arm through speech interaction using these quasi-symbols. These quasi symbols formed a hierarchical structure corresponding to the number of nodes in the model. The meaning of some of the quasi-symbols depended on the context, indicating that they are useful for human-robot interaction. © 2007 IEEE.

    DOI

  • Predicting object dynamics from visual images through active sensing experiences

    Shun Nishide, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation     2501 - 2506  2007年  [査読有り]

     概要を見る

    Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the learning phase, the authors use Recurrent Neural Network with Parametric Bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects, and robot motor values are input into a hierarchical neural network to link the static images to dynamic features (PB values). The neural network extracts prominent features that induce each object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated sequentially. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. Reducted grayscale images and shoulder pitch angles were input into the neural network to predict the dynamics of target objects. The results of the experiment proved that the technique is efficient for predicting the dynamics of the objects. © 2007 IEEE.

    DOI

  • Evaluation of two simultaneous continuous speech recognition with ICA BSS and MFT-based ASR

    Ryu Takeda, Shun'ichi Yamamoto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   4570 LNAI   384 - 394  2007年  [査読有り]

     概要を見る

    An adaptation of independent component analysis (ICA) and missing feature theory (MFT)-based ASR for two simultaneous continuous speech recognition is described. We have reported on the utility of a system with isolated word recognition, but the performance of the MFT-based ASR is affected by the configuration, such as an acoustic model. The system needs to be evaluated under a more general condition. It first separates the sound sources using ICA. Then, spectral distortion in the separated sounds is estimated to generate missing feature masks (MFMs). Finally, the separated sounds are recognized by MFT-based ASR. We estimate spectral distortion in the temporal-frequency domain in terms of feature vectors, and we generate MFMs. We tested an isolated word and the continuous speech recognition with a cepstral and spectral feature. The resulting system outperformed the baseline robot audition system by 13 and 6 points respectively on the spectral features. © Springer-Verlag Berlin Heidelberg 2007.

    DOI

  • Real-time auditory and visual talker tracking through integrating EM algorithm and particle filter

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   4570 LNAI   280 - 290  2007年  [査読有り]

     概要を見る

    This paper presents techniques that enable a talker tracking for effective human-robot interaction. We propose new way of integrating an EM algorithm and a particle filter to select an appropriate path for tracking the talker. It can easily adapt to new kinds of information for tracking the talker with our system. This is because our system estimates the position of the desired talker through means, variances, and weights calculated from EM training regardless of the numbers or kinds of information. In addition, to enhance a robot's ability to track a talker in real-world environments, we applied the particle filter to talker tracking after executing the EM algorithm. We also integrated a variety of auditory and visual information regarding sound localization, face localization, and the detection of lip movement. Moreover, we applied a sound classification function that allows our system to distinguish between voice, music, or noise. We also developed a vision module that can locate moving objects. © Springer-Verlag Berlin Heidelberg 2007.

    DOI

  • Auditory and visual integration based localization and tracking of multiple moving sounds in daily-life environments

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Workshop on Robot and Human Interactive Communication     399 - 404  2007年  [査読有り]

     概要を見る

    This paper presents techniques that enable talker tracking for effective human-robot interaction. To track moving people in daily-life environments, localizing multiple moving sounds is necessary so that robots can locate talkers. However, the conventional method requires an array of microphones and impulse response data. Therefore, we propose a way to integrate a cross-power spectrum phase analysis (CSP) method and an expectation-maximization (EM) algorithm. The CSP can localize sound sources using only two microphones and does not need impulse response data. Moreover, the EM algorithm increases the system's effectiveness and allows it to cope with multiple sound sources. We confirmed that the proposed method performs better than the conventional method. In addition, we added a particle filter to the tracking process to produce a reliable tracking path and the particle filter is able to integrate audio-visual information effectively. Furthermore, the applied particle filter is able to track people while dealing with various noises that are even loud sounds in the daily-life environments. ©2007 IEEE.

    DOI

  • Topic Estimation with Domain Extensibility for Guiding User”s Out-of-Grammar Utterance in Multi-Domain Spoken Dialogue Systems

    Satoshi IKEDA, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proceedings of International Conference on Spoken Language Processing (Interspeech-2007)   3   2057 - 2060  2007年  [査読有り]

  • Introducing utterance verification in spoken dialogue system to improve dynamic help generation for novice users

    Kazunori Komatani, Yuichiro Fukubayashi, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings of the 8th SIGdial Workshop on Discourse and Dialogue     202 - 205  2007年

     概要を見る

    A method is presented that helps novice users understand the language expressions that a system can accept, even from unacceptable utterances made that may contain automatic speech recognition errors. We have developed a method that dynamically generates help messages, which can avoid further unacceptable utterances from being made, by estimating a users' knowledge from their utterances. To improve the accuracy of the estimation, we developed a method to estimate a user's knowledge from utterance verification results. This method estimates whether a user knows an utterance pattern that the system considers acceptable, and suppresses useless help messages from being generated. © 2007 Association for Computational Linguistics.

  • Exploiting known sound source signals to improve ICA-based robot audition in speech separation and recognition

    Ryu Takeda, Kazuhiro Nakadai, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1757 - 1762  2007年  [査読有り]

     概要を見る

    This paper describes a new semi-blind source separation (semi-BSS) technique with independent component analysis (ICA) for enhancing a target source of interest and for suppressing other known interference sources. The semi-BSS technique is necessary for double-talk free robot audition systems in order to utilize known sound source signals such as self speech, music, or TV-sound, through a line-in or ubiquitous network. Unlike the conventional semi-BSS with ICA, we use the time-frequency domain convolution model to describe the reflection of the sound and a new mixing process of sounds for ICA. In other words, we consider that reflected sounds during some delay time are different from the original. ICA then separates the reflections as other interference sources. The model enables us to eliminate the frame size limitations of the frequency-domain ICA, and ICA can separate the known sources under a highly reverberative environment. Experimental results show that our method outperformed the conventional semi-BSS using ICA under simulated normal and highly reverberative environments. ©2007 IEEE.

    DOI

  • Two-way translation of compound sentences and arm motions by recurrent neural networks

    Tetsuya Ogata, Masamitsu Murase, Jim Tani, Kazunori Komatani, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1858 - 1863  2007年  [査読有り]

     概要を見る

    We present a connectionist model that combines motions and language based on the behavioral experiences of a real robot. Two models of recurrent neural network with parametric bias (RNNPB) were trained using motion sequences and linguistic sequences. These sequences were combined using their respective parameters so that the robot could handle many-to-many relationships between motion sequences and linguistic sequences. Motion sequences were articulated into some primitives corresponding to given linguistic sequences using the prediction error of the RNNPB model. The experimental task in which a humanoid robot moved its arm on a table demonstrated that the robot could generate a motion sequence corresponding to given linguistic sequence even if the motions or sequences were not included in the training data, and vice versa. ©2007 IEEE.

    DOI

  • Vocal imitation using physical vocal tract model

    Hisashi Kanda, Tetsuya Ogata, Kazunori Komatani, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1846 - 1851  2007年  [査読有り]

     概要を見る

    A vocal imitation system was developed using a computational model that supports the motor theory of speech perception. A critical problem in vocal imitation is how to generate speech sounds produced by adults, whose vocal tracts have physical properties (i.e., articulatory motions) differing from those of infants' vocal tracts. To solve this problem, a model based on the motor theory of speech perception, was constructed. This model suggests that infants simulate the speech generation by estimating their own articulatory motions in order to interpret the speech sounds of adults. Applying this model enables the vocal imitation system to estimate articulatory motions for unexperienced speech sounds that have not actually been generated by the system. The system was implemented by using Recurrent Neural Network with Parametric Bias (RNNPB) and a physical vocal tract model, called the Maeda model. Experimental results demonstrated that the system was sufficiently robust with respect to individual differences in speech sounds and could imitate unexperienced vowel sounds. ©2007 IEEE.

    DOI

  • A biped robot that keeps steps in time with musical beats while listening to music with its own ears

    Kazuyoshi Yoshii, Kazuhiro Nakadai, Toyotaka Torii, Yuji Hasegawa, Hiroshi Tsujino, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1743 - 1750  2007年  [査読有り]

     概要を見る

    We aim at enabling a biped robot to interact with humans through real-world music in daily-life environments, e.g., to autonomously keep its steps (stamps) in time with musical beats. To achieve this, the robot should be able to robustly predict the beat times in real time while listening to musical performance with its own ears (head-embedded microphones). However, this has not previously been addressed in most studies on music-synchronized robots due to the difficulty in predicting the beat times in real-world music. To solve this problem, we implemented a beat-tracking method developed in the field of music information processing. The predicted beat times are then used by a feedback-control method that adjusts the robot's step intervals to synchronize its steps in time with the beats. The experimental results show that the robot can adjust its steps in time with the beat times as the tempo changes. The resulting robot needed about 25 [s] to recognize the tempo change after it and then synchronize its steps. ©2007 IEEE.

    DOI

  • Discovery of other individuals by projecting a self-model through imitation

    Ryunosuke Yokoya, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1009 - 1014  2007年  [査読有り]

     概要を見る

    This paper proposes a novel model which enables a humanoid robot infant to discover other individual (e.g. human parent). In this work, the authors define "other individual" as an actor which can be predicted by a self-model. For modeling the developmental process of discovering ability, the following three approaches are employed. (i) Projection of a self-model for predicting other individual's actions. (ii) Mediation by a physical object between self and other individual. (iii) Introduction of infant imitation by parent. For creating the self-model of a robot, we apply Recurrent Neural Network with Parametric Bias (RNNPB) model which can learn the robot's body dynamics. For the other-model of a human, conventional hierarchical neural networks are attached to the RNNPB model as "conversion modules". Our target task is a moving an object. For evaluation of our model, human discovery experiments by the robot projecting its self-model were conducted. The results demonstrated that our method enabled the robot to predict the human's motions, and to estimate the human's position fairly accurately, which proved its adequacy. ©2007 IEEE.

    DOI

  • Auditory and visual integration based localization and tracking of humans in daily-life environments

    Hyun Don Kim, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     2021 - 2027  2007年  [査読有り]

     概要を見る

    The purpose of this research is to develop techniques that enable robots to choose and track a desired person for interaction in daily-life environments. Therefore, localizing multiple moving sounds and human faces is necessary so that robots can locate a desired person. For sound source localization, we used a cross-power spectrum phase analysis (CSP) method and showed that CSP can localize sound sources only using two microphones and does not need impulse response data. An expectation-maximization (EM) algorithm was shown to enable a robot to cope with multiple moving sound sources. For face localization, we developed a method that can reliably detect several faces using the skin color classification obtained by using the EM algorithm. To deal with a change in color state according to illumination condition and various skin colors, the robot can obtain new skin color features of faces detected by OpenCV, an open vision library, for detecting human faces. Finally, we developed a probability based method to integrate auditory and visual information and to produce a reliable tracking path in real time. Furthermore, the developed system chose and tracked people while dealing with various background noises that are considered loud, even in the daily-life environments. ©2007 IEEE.

    DOI

  • Design and implementation of a robot audition system for automatic speech recognition of simultaneous speech

    Shun'ichi Yamamoto, Kazuhiro Nakadai, Mikio Nakano, Hiroshi Tsujino, Jean Marc Valin, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2007 IEEE Workshop on Automatic Speech Recognition and Understanding, ASRU 2007, Proceedings     111 - 116  2007年  [査読有り]

     概要を見る

    This paper addresses robot audition that can cope with speech that has a low signal-to-noise ratio (SNR) in real time by using robot-embedded microphones. To cope with such a noise, we exploited two key ideas; Preprocessing consisting of sound source localization and separation with a microphone array, and system integration based on missing feature theory (MFT). Preprocessing improves the SNR of a target sound signal using geometric source separation with multichannel post-filter. MFT uses only reliable acoustic features in speech recognition and masks unreliable parts caused by errors in preprocessing. MFT thus provides smooth integration between preprocessing and automatic speech recognition. A real-time robot audition system based on these two key ideas is constructed for Honda ASIMO and Humanoid SIG2 with 8-ch microphone arrays. The paper also reports the improvement of ASR performance by using two and three simultaneous speech signals. © 2007 IEEE.

    DOI

  • 多重奏を対象とした音源同定:混合音テンプレートを用いた音の重なりに頑健な特徴量への重みづけおよび音楽的文脈の利用

    北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    電子情報通信学会論文誌   J89-D ( 12 ) 2721 - 2733  2006年12月

  • 隠れ状態を有する連続な状態空間での強化学習法の提案

    鈴木貴晴, 有江浩明, 尾形哲也, 谷淳, 菅野重樹

    第7回システムインテグレーション部門講演会 (SI2006), 2C2-7, 計測自動制御学会    2006年12月

  • 自己組織化回路素子SONEへの教師あり学習機能の付与

    金天海, 尾形哲也, 菅野重樹

    第7回システムインテグレーション部門講演会 (SI2006), 2C2-6, 計測自動制御学会    2006年12月

  • 自己組織化回路素子SONEにおけるフリップフロップ素子の導入によるシーケンスの分節化と統合

    出澤純一, 金天海, 尾形哲也, 菅野重樹

    第7回システムインテグレーション部門講演会 (SI2006), 2C2-5, 計測自動制御学会    2006年12月

  • IECを用いた適応的なインタラクションシステムの実現

    小林大三, 遠藤ちひろ, 松本猛, 菅佑樹, 尾形哲也, 菅野重樹

    第7回システムインテグレーション部門講演会 (SI2006), 1L3-5, 計測自動制御学会    2006年12月

  • 能動知覚経験に基づく物体静止画像からの挙動推定

    西出俊, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第7回システムインテグレーション部門講演会 (SI2006), 1B2-4, 計測自動制御学会    2006年12月

  • 視聴覚情報統合及びEMアルゴリズムを用いた人物追跡システム実現

    Hyun-Don Kim, 駒谷和範, 尾形哲也, 奥乃博

    第24回 AI チャレンジ研究会, 人工知能学会   SIG-Challenge-0624-8   51 - 58  2006年11月

  • ICAによる音源分離とMFTに基づく音声認識の同時発話認識による評価,

    武田龍, 山本俊一, 駒谷和範, 尾形哲也, 奥乃博

    第24回 AI チャレンジ研究会, 人工知能学会   SIG-Challenge-0624-2   9 - 16  2006年11月

  • RNNPBによる視聴覚情報変換を利用したロボットの身体・音声表現

    尾形哲也, 小嶋秀樹, 駒谷和範, 奥乃博

    言語理解とコミュニケーション研究会他, TL2006-22 NLC2006-18 PRMU2006-99 pp.27-32(TL), pp.27-32(NLC), pp.45-50(PRMU), 電子情報通信学会    2006年10月

  • Generation of Robot Motions from Environmental Sounds using Inter-modality Mapping by RNNPB

    Tetsuya OGATA, Yuya HATTORI, Hideki KOZIMA, Kazunori KOMATANI, Hiroshi G. OKUNO

    Proc. of International Workshop on Epigenetic Robotics     95 - 102  2006年09月

  • パラメータ最適化による実環境同時発話認識向上とそのオンライン処理の実装

    山本俊一, 中臺一博, 中野幹生, 辻野広司, JEAN-MARC VALIN, 駒谷和範, 尾形哲也, 奥乃博

    第24回日本ロボット学会学術講演    2006年09月

  • ICAとミッシングフィーチャマスク自動生成によるロボット聴覚

    武田龍, 山本俊一, 駒谷和範, 尾形哲也, 奥乃博

    第24回日本ロボット学会学術講演会    2006年09月

  • CTRNNを用いた強化学習法による連続な行動出力の獲得

    有江浩明, 尾形哲也, 谷淳, 菅野重樹

    第24回日本ロボット学会学術講演会    2006年09月

  • リカレントニューラルネットワークによるロボットの異種感覚モダリティ変換

    尾形哲也, 服部佑哉, 小嶋秀樹, 駒谷和範, 奥乃博

    第24回日本ロボット学会学術講演会    2006年09月

  • ユーザの変化する主観に対応するコミュニケーションロボットの行動獲得

    松本猛, 遠藤ちひろ, 小林大三, 菅佑樹, 尾形哲也, 菅野重樹

    第24回日本ロボット学会学術講演会    2006年09月

  • 人間ロボット協調のためのRNNPBによる疑似シンボルの獲得とその階層性の解析

    村瀬昌満, 松本祥平, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第24回日本ロボット学会学術講演会    2006年09月

  • ロボットの身体経験に基づくRNNPBを用いた模倣動作の自律的獲得

    横矢龍之介, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    第24回日本ロボット学会学術講演会    2006年09月

  • 自己組織化論理回路における対ノイズ性能の向上

    金天海, 出澤純一, 尾形哲也, 菅野重樹

    第24回日本ロボット学会学術講演会    2006年09月

  • 人間とプログラムのハイブリッド評価を用いた対話型進化的計算コミュニケーションロボットの行動獲得への応用

    菅佑樹, 尾形哲也, 菅野重樹

    第24回日本ロボット学会学術講演会    2006年09月

  • 可聴音波を用いたAH法による遮断された物体の検出と距離計測

    丹羽治彦, 尾形哲也, 駒谷和範, 奥乃博

    日本音響学会応用音響研究会    2006年09月

  • ユーザの評価と音響的特徴との確率的統合に基づくハイブリッド型楽曲推薦システム

    吉井和佳, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    第66回音楽情報処理研究会, 2006-MUS, 情報処理学会    2006年08月

  • Instrogram: 発音時刻検出とF0推定の不要な楽器音認識手法

    北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    第66回音楽情報処理研究会, 2006-MUS, 情報処理学会    2006年08月

  • 伴奏音抑制と高信頼度フレーム選択に基づく楽曲中の歌声の歌手名同定手法

    藤原弘将, 北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    第66回音楽情報処理研究会, 2006-MUS, 情報処理学会    2006年08月

  • 音声対話システムにおける発話パターンを教示するヘルプの動的生成

    福林雄一郎, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会研究会資料SIG-SLUD-A601-03 人工知能学会     13 - 18  2006年07月

  • 伴奏音抑制と高信頼度フレーム選択に基づく楽曲の歌手名同定手法

    藤原弘将, 北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   47 ( 6 ) 1831 - 1843  2006年06月

  • データベース検索における対話文脈を利用した音声言語理解

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会論文誌   47 ( 6 ) 1802 - 1811  2006年06月

  • Robot Imitation from Active-Sensing Experiences

    Ryunosuke YOKOYA, Tetsuya OGATA, Jun TANI, Kazunori KOMATANI, Hiroshi G. OKUNO

    International Conference on Development and Learning (ICDL 2006)    2006年06月

  • 遺伝的アルゴリズムを用いたパラメータ最適化による話者位置に基づく同時発話認識の向上

    山本 俊一, 中臺 一博, 中野 幹生, 辻野 広司, VALIN Jean-Marc, 武田 龍, 駒谷 和範, 尾形 哲也, 奥乃 博

    ヒューマンインタフェース学会論文誌   8 ( 2 ) 203 - 212  2006年05月

    CiNii

  • ドラムパートのリアルタイム編集機能付きオーディオプレイヤー

    吉井和佳, 後藤真考, 駒谷和範, 尾形哲也, 奥乃博

    インタラクション2006, インタラクティブセッション, 情報処理学会    2006年05月

  • IECを用いたコミュニケーションロボットにおける人間の主観変化への適応

    遠藤ちひろ, 小林大三, 松本猛, 菅佑樹, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会    2006年05月

  • 人間とのコミュニケーションを通したロボットの内分泌系モデルと情動表出の段階的進化

    菅佑樹, 高田広太郎, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会    2006年05月

  • 自己組織化ネットワーク素子群における対ノイズ性能向上

    出澤純一, 金天海, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会    2006年05月

  • CTRNNを用いた連続な状態空間における強化学習法の提案

    有江浩明, 尾形哲也, 谷淳, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会    2006年05月

  • ローカルルールに基づいた理論回路の自己組織アルゴリズム

    金天海, 尾形哲也, 菅野重樹

    計測自動制御学会論文誌   42 ( 4 ) 334 - 341  2006年04月

  • 多重奏中特定パートの自動採譜における複数特徴量の自動重み付け

    糸山克寿, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 表題音楽アノテーションのための階層的物語タグの設計

    西山正紘, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 擬音語表現を利用した環境音のためのXMLタグの設計と自動付与

    田口明裕, 北原鉄朗, 石原一志, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • GAによる話者位置への同時発話認識システムの最適化

    山本俊一, 中臺一博, 中野幹生, 辻野広司, Jean-Marc VALIN, 武田龍, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 音源定位及び唇の動き検出による複数ユーザ環境における発話者認識

    Hyun-Don KIM, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 強化学習による人間位置に基づいたロボットの挙動選択

    田崎豪, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 複数ドメイン音声対話システムにおける対話履歴を利用したドメイン選択の高精度化

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 音声対話システムにおけるユーザの誤り原因の推定に基づく動的ヘルプ生成

    福林雄一朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • ICAによる音源分離とミッシングフィーチャーマスクによる同時発話認識

    武田龍, 山本俊一, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • RNNPBを用いたモダリティ間マッピングによるロボットの動作生成

    服部佑哉, 駒谷和範, 尾形哲也, 小嶋秀樹, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • RNNPBを用いて獲得した疑似シンボルによる人間とロボットの協調の実現

    松本祥平, 駒谷和範, 尾形哲也, 谷淳, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • アクティブセンシングを用いたロボットによる模倣動作の自律的獲得

    横矢龍之介, 駒谷和範, 尾形哲也, 谷淳, 奥乃博

    情報処理学会第68回全国大会    2006年03月

  • 調波構造抽出と高信頼度フレーム選択を用いた雑音下での話者識別

    藤原弘将, 北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    日本音響学会春期研究発表会    2006年03月

  • Instrogram: 楽器存在確率に基づく音楽視覚表現法

    北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    日本音響学会春期研究発表会    2006年03月

  • ゲーム理論に基づく中心化理論の再定式化と日英コーパスを用いた統計的検証

    白松俊, 駒谷和範, 尾形哲也, 橋田浩一, 奥乃博

    言語処理学会第12回年次大会    2006年03月

  • An error correction framework based on drum pattern periodicity for improving drum sound detection

    Kazuyoshi Yoshii, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13     5095 - 5098  2006年  [査読有り]

     概要を見る

    This paper presents a framework for correcting errors of automatic drum sound detection focusing on the periodicity of drum patterns. We define drum patterns as periodic structures found in onset sequences of bass and snare drum sounds. Our framework extracts periodic drum patterns from imperfect onset sequences of detected drum sounds (bottom-up processing) and corrects errors using the periodicity of the drum patterns (top-down processing). We implemented this framework on our drum-sound detection system. We first obtained onset sequences of the drum sounds with our system and extracted drum patterns. On the basis of our observation that the same drum patterns tend to be repeated, we detected time points which deviate from the periodicity as error candidates. Finally, we verified each error candidate to judge whether it is an actual onset or not. Experiments of drum sound detection for polyphonic audio signals of popular CD recordings showed that our correction framework improved the average detection accuracy from 77.4% to 80.7%.

  • Instrogram: A new musical instrument recognition technique without using onset detection nor F0 estimation

    Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings   5   229 - 232  2006年  [査読有り]

     概要を見る

    This paper describes a new technique for recognizing musical instruments in polyphonic music. Because the conventional framework for musical instrument recognition in polyphonic music had to estimate the onset time and fundamental frequency (F0) of each note, instrument recognition strictly suffered from errors of onset detection and F0 estimation. Unlike such a note-based processing framework, our technique calculates the temporal trajectory of instrument existence probabilities for every possible F0, and the results are visualized with a spectrogram-like graphical representation called instrogram. The instrument existence probability is defined as the product of a nonspecific instrument existence probability calculated using PreFEst and a conditional instrument existence probability calculated using the hidden Markov model. Experimental results show that the obtained instrograms reflect the actual instrumentations and facilitate instrument recognition. © 2006 IEEE.

    DOI

  • F0 estimation method for singing voice in polyphonic audio signal based on statistical vocal model and viterbi search

    Hiromasa Fujihara, Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings   5   253 - 256  2006年  [査読有り]

     概要を見る

    This paper describes a method for estimating F0s of vocal from polyphonic audio signals. Because melody is sung by a singer in many musical pieces, the estimation of F0s of the vocal part is useful for many applications. Based on existing multiple-F0 estimation method, we evaluate the vocal probabilities of the harmonic structure of each F0 candidate. In order to calculate the vocal probabilities of the harmonic structure, we extract and resynthesize the harmonic structure by using a sinusoidal model and extract feature vectors. Then, we evaluate the vocal probability by using vocal and non-vocal Gaussian mixture models (GMMs). Finally, we track F0 trajectories using these probabilities based on Viterbi search. Experimental results show that our method improves estimation accuracy from 78.1% to 84.3%, which is 28.3% reduction of misestimation. © 2006 IEEE.

    DOI

  • Human-Robot Interaction System Using Interactive EC

    Y. Suga, C. Endo, T. Ogata, S. Sugano

    IEEE/RSJ International conference on Intelligent Robots and Systems     145  2006年

    DOI

  • Leak energy based missing feature mask generation for ICA and GSS and its evaluation with simultaneous speech recognition.

    Shun'ichi Yamamoto, Ryu Takeda, Kazuhiro Nakadai, Mikio Nakano, Hiroshi Tsujino, Jean-Marc Valin, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proc. of ISCA Tutorial and Research Workshop on Statistical and Perceptual Audition (SAPA2006)     42 - 47  2006年

  • Missing-feature based speech recognition for two simultaneous speech signals separated by ICA with a pair of humanoid ears

    Ryu Takeda, Shun'ichi Yamamoto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     878 - 885  2006年  [査読有り]

     概要を見る

    Robot audition is a critical technology in making robots symbiosis with people. Since we hear a mixture of sounds in our daily lives, sound source localization and separation, and recognition of separated sounds are three essential capabilities. Sound source localization has been recently studied well for robots, while the other capabilities still need extensive studies. This paper reports the robot audition system with a pair of omni-directional microphones embedded in a humanoid to recognize two simultaneous talkers. It first separates sound sources by Independent Component Analysis (ICA) with single-input multiple-output (SIMO) model. Then, spectral distortion for separated sounds is estimated to identify reliable and unreliable components of the spectrogram. This estimation generates the missing feature masks as spectrographic masks. These masks are then used to avoid influences caused by spectral distortion in automatic speech recognition based on missing-feature method. The novel ideas of our system reside in estimates of spectral distortion of temporal-frequency domain in terms of feature vectors. In addition, we point out that the voice-activity detection (VAD) is effective to overcome the weak point of ICA against the changing number of talkers. The resulting system outperformed the baseline robot audition system by 15 %. © 2006 IEEE.

    DOI

  • Dynamic perception after visually guided grasping by a human-like autonomous robot

    Mototaka Suzuki, Kuniaki Noda, Yuki Suga, Tetsuya Ogata, Shigeki Sugano

    Advanced Robotics   20 ( 2 ) 233 - 254  2006年  [査読有り]

     概要を見る

    We will explore dynamic perception following the visually guided grasping of several objects by a human-like autonomous robot. This competency serves for object categorization. Physical interaction with the hand-held object gives the neural network of the robot the rich, coherent and multi-modal sensory input. Multi-layered self-organizing maps are designed and examined in static and dynamic conditions. The results of the tests in the former condition show its capability of robust categorization against noise. The network also shows better performance than a single-layered map does. In the latter condition we focus on shaking behavior by moving only the forearm of the robot. In some combinations of grasping style and shaking radius the network is capable of categorizing two objects robustly. The results show that the network capability to achieve the task largely depends on how to grasp and how to move the objects. These results together with a preliminary simulation are promising toward the self-organization of a high degree of autonomous dynamic object categorization. © 2006 Taylor & Francis Group, LLC.

    DOI

  • 可聴域音波の広指向性と回折を利用した同時複数物体定位と遮蔽物検出

    丹羽治彦, 尾形哲也, 駒谷和範, 奥乃博

    電子情報通信学会超音波研究会    2006年01月

  • Multi-domain spoken dialogue system with extensibility and robustness against speech recognition errors

    Kazunori Komatani, Naoyuki Kanda, Mikio Nakano, Kazuhiro Nakadai, Hiroshi Tsujino, Tetsuya Ogata, Hiroshi G. Okuno

    COLING/ACL 2006 - SIGdial06: 7th SIGdial Workshop on Discourse and Dialogue, Proceedings of the Workshop     9 - 17  2006年

     概要を見る

    We developed a multi-domain spoken dialogue system that can handle user requests across multiple domains. Such systems need to satisfy two requirements: extensibility and robustness against speech recognition errors. Extensibility is required to allow for the modification and addition of domains independent of other domains. Robustness against speech recognition errors is required because such errors are inevitable in speech recognition. However, the systems should still behave appropriately, even when their inputs are erroneous. Our system was constructed on an extensible architecture and is equipped with a robust and extensible domain selection method. Domain selection was based on three choices: (I) the previous domain, (II) the domain in which the speech recognition result can be accepted with the highest recognition score, and (III) other domains. With the third choice we newly introduced, our system can prevent dialogues from continuously being stuck in an erroneous domain. Our experimental results, obtained with 10 subjects, showed that our method reduced the domain selection errors by 18.3%, compared to a conventional method. © 2006 Association for Computational Linguistics.

    DOI

  • Instrogram: A new musical instrument recognition technique without using onset detection nor F0 estimation

    Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13     5087 - 5090  2006年  [査読有り]

     概要を見る

    This paper describes a new technique for recognizing musical instruments in polyphonic music. Because the conventional framework for musical instrument recognition in polyphonic music had to estimate the onset time and fundamental frequency (FO) of each note, instrument recognition strictly suffered from errors of onset detection and FO estimation. Unlike such a note-based processing framework, our technique calculates the temporal trajectory of instrument existence probabilities for every possible FO, and the results are visualized with a spectrogram-like graphical representation called instrogram. The instrument existence probability is defined as the product of a nonspecific instrument existence probability calculated using PreFEst and a conditional instrument existence probability calculated using the hidden Markov model. Experimental results show that the obtained instrograms reflect the actual instrumentations and facilitate instrument recognition.

  • F0 estimation method for singing voice in polyphonic audio signal based on statistical vocal model and Viterbi search

    Hiromasa Fujihara, Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2006 IEEE International Conference on Acoustics, Speech and Signal Processing, Vols 1-13     5111 - 5114  2006年  [査読有り]

     概要を見る

    This paper describes a method for estimating FOs of vocal from polyphonic audio signals. Because melody is sung by a singer in many musical pieces, the estimation of FOs of the vocal part is useful for many applications. Based on existing multiple-170 estimation method, we evaluate the vocal probabilities of the harmonic structure of each FO candidate. In order to calculate the vocal probabilities of the harmonic structure, we extract and resynthesize the harmonic structure by using a sinusoidal model and extract feature vectors. Then, we evaluate the vocal probability by using vocal and non-vocal Gaussian mixture models (GMMs). Finally, we track FO trajectories using these probabilities based on Viterbi search. Experimental results show that our method improves estimation accuracy from 78.1% to 84.3%, which is 28.3% reduction of misestimation.

  • An error correction framework based on drum pattern periodicity for improving drum sound detection

    Kazuyoshi Yoshii, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings   5   237 - 240  2006年  [査読有り]

     概要を見る

    This paper presents a framework for correcting errors of automatic drum sound detection focusing on the periodicity of drum patterns. We define drum patterns as periodic structures found in onset sequences of bass and snare drum sounds. Our framework extracts periodic drum patterns from imperfect onset sequences of detected drum sounds (bottom-up processing) and corrects errors using the periodicity of the drum patterns (top-down processing). We implemented this framework on our drum-sound detection system. We first obtained onset sequences of the drum sounds with our system and extracted drum patterns. On the basis of our observation that the same drum patterns tend to be repeated, we detected time points which deviate from the periodicity as error candidates. Finally, we verified each error candidate to judge whether it is an actual onset or not. Experiments of drum sound detection for polyphonic audio signals of popular CD recordings showed that our correction framework improved the average detection accuracy from 77.4% to 80.7%. © 2006 IEEE.

    DOI

  • Genetic Algorithm based Improvement of Robot”s Hearing Capabilities in Separating and Recognizing Simultaneous Speech Signals

    Shun'ichi YAMAMOTO, Kazuhiro NAKADAI, Miko NAKANO, Hiroshi TSUJINO, Jean-Marc Varin, Ryu TAKEDA, Tetsuya OGATA, Kazunori KOMATANI, Hiroshi G. OKUNO

    Nineteenth International Conference on Industrial, Engineering and Other Applications of Applied Intelligence Systems (IEA/AIE-2006)   4031 LNAI   207 - 217  2006年  [査読有り]

    DOI

  • Recognition of simultaneous speech by estimating reliability of separated signals for robot audition

    Shun'ichi Yamamoto, Ryu Takeda, Kazuhiro Nakadai, Mikio Nakano, Hiroshi Tsujino, Jean Marc Valin, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   4099 LNAI   484 - 494  2006年  [査読有り]

     概要を見る

    "Listening to several things at once" is a people's dream and one goal of AI and robot audition, because people can listen to at most two things at once according to psychophysical observations. Current noise reduction techniques cannot help to achieve this goal because they assume quasi-stationary noises, not interfering speech signals. Since robots are used in various environments, robot audition systems require minimum a priori information about their acoustic environments and speakers. We evaluate a missing feature theory approach that interfaces between sound source separation (SSS) and automatic speech recognition. The essential part is the estimate of reliability of each feature of separated sounds. We tested two kinds of robot audition systems that use SSS: independent component analysis (ICA) with two microphones, and geometric source separation (GSS) with eight microphones. For each SSS, automatic missing feature mask generation is developed. The recognition accuracy of two simultaneous speech improved to an average of 67.8 and 88.0% for ICA and GSS, respectively. © Springer-Verlag Berlin Heidelberg 2006.

    DOI

  • Improvement against noises in self-organizing logic circuit

    Chyon Hae Kim, Jyun Ichi Idesawa, Shigeki Sugano, Tetsuya Ogata

    Proceedings of IEEE ICIA 2006 - 2006 IEEE International Conference on Information Acquisition     53 - 58  2006年  [査読有り]

     概要を見る

    We proposed and evaluated a network learning method called self-organizing network elements (SONE). Autonomous exploration of effective output, the use of simple external parameters, and low calculation costs were functions achieved for a robot system with this method. However, there is the need to improve performance against noises for learning more complicated tasks. Therefore, we propose a technique to adjust thresholds in a self-organizing logic circuit based on SONE. In our experiments, performing 3-bit operation tasks, controlling a Khepera robot, and generating new elements was appropriately controlled with this technique. Also, network performance against noises was improved as a result of using the proposed method. ©2006 IEEE.

    DOI

  • Reinforcement learning algorithm with CTRNN in continuous action space

    Hiroaki Arie, Jun Namikawa, Tetsuya Ogata, Jun Tani, Shigeki Sugano

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   4232 LNCS   387 - 396  2006年  [査読有り]

     概要を見る

    There are some difficulties in applying traditional reinforcement learning algorithms to motion control tasks of robot. Because most algorithms are concerned with discrete actions and based on the assumption of complete observability of the state. This paper deals with these two problems by combining the reinforcement learning algorithm and CTRNN learning algorithm. We carried out an experiment on the pendulum swing-up task without rotational speed information. It is shown that the information about the rotational speed, which is considered as a hidden state, is estimated and encoded on the activation of a context neuron. As a result, this task is accomplished in several hundred trials using the proposed algorithm. © Springer-Verlag Berlin Heidelberg 2006.

    DOI

  • Adaptive human-robot interaction system using interactive EC

    Yuki Suga, Chihiro Endo, Daizo Kobayashi, Takeshi Matsumoto, Shigeki Sugano, Tetsuya Ogata

    IEEE International Conference on Intelligent Robots and Systems     3663 - 3668  2006年  [査読有り]

     概要を見る

    We created a human-robot communication system that can adapt to user preferences that can easily change through communication. Even if any learning algorithms are used, evaluating the human-robot interaction is indispensable and difficult. To solve this problem, we installed a machine-learning algorithm called Interactive Evolutionary Computation (IEC) into a communication robot named WAMOEBA-3. IEC is a kind of evolutionary computation like a genetic algorithm. With IEC, the fitness function is performed by each user. We carried out experiments on the communication learning system using an advanced IEC system named HMHE. Before the experiments, we did not tell the subjects anything about the robot, so the interaction differed among the experimental subjects. We could observe mutual adaptation, because some subjects noticed the robot's functions and changed their interaction. From the results, we confirmed that, in spite of the changes of the preferences, the system can adapt to the interaction of multiple users. © 2006 IEEE.

    DOI

  • Efficient organization of network topology based on reinforcement signals

    Chyon Hae Kim, Shigeki Sugano, Tetsuya Ogata

    IEEE International Conference on Intelligent Robots and Systems     3154 - 3159  2006年  [査読有り]

     概要を見る

    We developed a learning system for autonomous robots that allows for autonomous exploration of the effective output, and has simple external parameters and a low calculation cost. We propose the concept of self-organizing network elements (SONE) for creating learning systems with these characteristics. We created and evaluated a self-organizing logic circuit by using this concept. Our results indicated this learning system had the characteristics. © 2006 IEEE.

    DOI

  • Real-time robot audition system that recognizes simultaneous speech in the real world

    Shun'ichi Yamamoto, Kazuhiro Nakadai, Mikio Nakano, Hiroshi Tsujino, Jean Marc Valin, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     5333 - 5338  2006年  [査読有り]

     概要を見る

    This paper presents a robot audition system that recognizes simultaneous speech in the real world by using robot-embedded microphones. We have previously reported Missing Feature Theory (MFT) based integration of Sound Source Separation (SSS) and Automatic Speech Recognition (ASR) for building robust robot audition. We demonstrated that a MFT-based prototype system drastically improved the performance of speech recognition even when three speakers talked to a robot simultaneously. However, the prototype system had three problems; being offline, hand-tuning of system parameters, and failure in Voice Activity Detection (VAD). To attain online processing, we introduced FlowDesigner-based architecture to integrate sound source localization (SSL), SSS and ASR. This architecture brings fast processing and easy implementation because it provides a simple framework of shared-object-based integration. To optimize the parameters, we developed Genetic Algorithm (GA) based parameter optimization, because it is difficult to build an analytical optimization model for mutually dependent system parameters. To improve VAD, we integrated new VAD based on a power spectrum and location of a sound source into the system, since conventional VAD relying only on power often fails due to low signal-to-noise ratio of simultaneous speech. We, then, constructed a robot audition system for Honda ASIMO. As a result, we showed that the system worked online and fast, and had a better performance in robustness and accuracy through experiments on recognition of simultaneous speech in a noisy and echoic environment. © 2006 IEEE.

    DOI

  • Multiple acoustical holography method for localization of objects in broad range using audible sound

    Haruhiko Niwa, Tetsuya Ogata, Kazunori Komatani, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     1145 - 1150  2006年  [査読有り]

     概要を見る

    This paper describes a new acoustic localization method using audible sound, which can be applied over a broader range of search directions. In the field of Robotics, most conventional indoor localization systems based on sonar range finders use ultrasound to obtain a highly accurate distance. Because ultrasound has high directivity, many measurements are required to localize objects in a large space. To achieve localization with one-time measurement, we use audible sounds. We then calculate an intensity field of the reflection sound to estimate object positions. Although acoustical holography (AH) is a well-known technique to do this, it has problems in that it generates false images. We propose multiple AH (MAH) to solve this problem. The method is used to divide a measurement plane into sub-planes and to apply AH to each sub-plane. By integrating the results of applying AH to the sub-planes, false images can be suppressed because the positions of the false images differ depending on the position of the sub-plane. In addition, we use multiple frequencies to advance an accuracy of the localization based on MAH in a real environment. We constructed a localization system with only one speaker and a microphone array. In both simulation and actual experiments, we confirmed that MAH was effective method for the suppression of false image and could be used within the range of an angle view of 120 deg. ©2006 IEEE.

    DOI

  • Experience based imitation using RNNPB

    Ryunosuke Yokoya, Tetsuya Ogata, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    IEEE International Conference on Intelligent Robots and Systems     3669 - 3674  2006年  [査読有り]

     概要を見る

    Robot imitation is a useful and promising alternative to robot programming. Robot imitation involves two crucial issues. The first is how a robot can imitate a human whose physical structure and properties differ greatly from its own. The second is how the robot can generate various motions from finite programmable patterns (generalization). This paper describes a novel approach to robot imitation based on its own physical experiences. Let us consider a target task of moving an object on a table. For imitation, we focused on an active sensing process in which the robot acquires the relation between the object's motion and its own arm motion. For generalization, we applied a recurrent neural network with parametric bias (RNNPB) model to enable recognition/generation of imitation motions. The robot associates the arm motion which reproduces the observed object's motion presented by a human operator. Experimental results demonstrated that our method enabled the robot to imitate not only motion it has experienced but also unknown motion, which proved its capability for generalization. © 2006 IEEE.

    DOI

  • Dynamic Help Generation by Estimating User”s Mental Model in Spoken Dialogue Systems

    Yuichiro FUKUBAYASHI, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proc. of International Conference on Spoken Language Processing (Interspeech-2006)   4   1946 - 1949  2006年  [査読有り]

  • Improving speech recognition of two simultaneous speech signals by integrating ICA BSS and automatic missing feature mask generation

    Rya Takeda, Shu N.Ichi Yamamoto, Kazunori Komatoni, Tetsuya Ogata, Hiroshi G. Okuno

    INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP   5   2302 - 2305  2006年  [査読有り]

     概要を見る

    Robot audition systems require capabilities for sound source separation and the recognition of separated sounds, since we hear a mixture of sounds in our daily lives, especially mixed of speech. We report a robot audition system with a pair of omni-directional microphones embedded in a humanoid that recognizes two simultaneous talkers. It first separates the sound sources by Independent Component Analysis (ICA) with the single-input multiple-output (SIMO) model. Then, spectral distortion in the separated sounds is then estimated to generate missing feature masks. Finally, the separated sounds are recognized by missing-feature theory (MFT) for Automatic Speech Recognition (ASR). The novel aspects of our system involve estimates of spectral distortion in the temporal-frequency domain in terms of feature vectors and based on estimates error in SIMO-ICA signals. The resulting system outperformed the baseline robot audition system by 7 %.

  • Speaker identification under noisy environments by using harmonic structure extraction and reliable frame weighting

    Hiromasa Fujihara, Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    INTERSPEECH 2006 and 9th International Conference on Spoken Language Processing, INTERSPEECH 2006 - ICSLP   3   1459 - 1462  2006年  [査読有り]

     概要を見る

    We present methods for automatic speaker identification in noisy environments. To improve noise robustness of speaker identification, we developed two methods, the harmonic structure extraction method and the reliable frame weighting method. The harmonic structure extraction method enables the speaker of input speech signals to be identified after environmental noise has been reduced. This method first extracts harmonic components of the speech from the sound mixtures and then resynthesizes a clean speech signal by using a sinusoidal model driven by harmonic components. The reliable frame weighting method then determines how each frame of the resynthesized speech is reliable (i.e. little influenced by environmental noises) by using two Gaussian mixture models for the speech and noise. The speaker can be robustly identified by attaching importance to reliable frames. Experimental results with thirty speakers showed that our method was able to reduce the influences of environmental noise and achieved an error rate of 10.7%, while the error rate for a conventional method was 18.9%.

  • Automatic feature weighting in automatic transcription of specified part in polyphonic music

    Katsutoshi Itoyama, Tetsuro Kitahara, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISMIR 2006 - 7th International Conference on Music Information Retrieval     172 - 175  2006年

     概要を見る

    We studied the problem of automatic music transcription (AMT) for polyphonic music. AMT is an important task for music information retrieval because AMT results enable retrieving musical pieces, high-level annotation, demixing, etc. We attempted to transcribe a part played by an instrument specified by users (specified part tracking). Only two timbre models are required in the specified part tracking to identify the specified musical instrument even when the number of instruments increases. This transcription is formulated into a time-series classification problem with multiple features. We furthermore attempted to automatically estimate weights of the features, because the importance of these features varies for each musical signal. We estimated quasi-optimal weights of the features using a genetic algorithm for each musical signal. We tested our AMT system using trio stereo musical signals. Accuracies with our feature weighting method were 69.8% on average, whereas those without feature weighting were 66.0%. © 2006 University of Victoria.

  • Hybrid collaborative and content-based music recommendation using probabilistic model with latent user preferences

    Kazuyoshi Yoshii, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISMIR 2006 - 7th International Conference on Music Information Retrieval     296 - 301  2006年

     概要を見る

    This paper presents a hybrid music recommendation method that solves problems of two prominent conventional methods: collaborative filtering and content-based recommendation. The former cannot recommend musical pieces that have no ratings because recommendations are based on actual user ratings. In addition, artist variety in recommended pieces tends to be poor. The latter, which recommends musical pieces that are similar to users' favorites in terms of music content, has not been fully investigated. This induces unreliability in modeling of user preferences; the content similarity does not completely reflect the preferences. Our method integrates both rating and content data by using a Bayesian network called an aspect model. Unobservable user preferences are directly represented by introducing latent variables, which are statistically estimated. To verify our method, we conducted experiments by using actual audio signals of Japanese songs and the corresponding rating data collected from Amazon. The results showed that our method outperforms the two conventional methods in terms of recommendation accuracy and artist variety and can reasonably recommend pieces even if they have no ratings. © 2006 University of Victoria.

  • Evolutionary approach for designing the behavior generator of communication robot

    Yuki Suga, Tetsuya Ogata, Shigeki Sugano

    2006 SICE-ICASE International Joint Conference     2104 - 2109  2006年  [査読有り]

     概要を見る

    Our goal is to create the robot system which interacts with human users keeping their interest during a long period. We focus on the Interactive Evolutionary Computation (IEC) technique to achieve this goal. Although the IEC enables users to design various systems which reflect their subjective preferences, it forces users to evaluate a huge number of individuals in the genetic pool during the evolution period. To solve this problem, we propose a refined IEC technique, named Human-Machine Hybrid Evaluation (HMHE), which selects the representative genes for user evaluation and estimates the evaluation results of the other genes. It can increase the population size without increasing the users' evaluation processes. We carried out some simulations where a humanoid robot with our method interacted with a user. The experimental results demonstrated that the HMHE could continue to generate the various robot behaviors by adapting to the transition of user's subjective preferences. © 2006 ICASE.

    DOI

  • Automatic synchronization between lyrics and music CD recordings based on Viterbi alignment of segregated vocal signals

    Hiromasa Fujihara, Masataka Goto, Jun Ogata, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISM 2006 - 8th IEEE International Symposium on Multimedia     257 - 264  2006年  [査読有り]

     概要を見る

    This paper describes a system that can automatically synchronize between polyphonic musical audio signals and corresponding lyrics. Although there were methods that can synchronize between monophonie speech signals and corresponding text transcriptions by using Viterbi alignment techniques, they cannot be applied to vocals in CD recordings because accompaniment sounds often overlap with vocals. To align lyrics with such vocals, we therefore developed three methods: a method for segregating vocals from polyphonic sound mixtures, a method for detecting vocal sections, and a method for adapting a speech-recognizer phone model to segregated vocal signals. Experimental results for 10 Japanese popular-music songs showed that our system can synchronize between music and lyrics with satisfactory accuracy for 8 songs. © 2006 IEEE.

    DOI

  • Musical instrument recognizer "instrogram" and its application to music retrieval based on instrumentation similarity

    Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISM 2006 - 8th IEEE International Symposium on Multimedia     265 - 272  2006年  [査読有り]

     概要を見る

    Instrumentation is an important cue in retrieving musical content. Conventional methods for instrument recognition performing notewise require accurate estimation of the onset time and fundamental frequency (F0) for each note, which is not easy in polyphonic music. This paper presents a non-notewise method for instrument recognition in polyphonic musical audio signals. Instead of such note-wise estimation, our method calculates the temporal trajectory of instrument existence probabilities for every F0 and visualizes it as a spectrogram-like graphical representation, called an instrogram. This method can avoid the influence by errors of onset detection and F0 estimation because it does not use them. We also present methods for MPEG-7-based instrument annotation and music information retrieval based on the similarity between instrograms. Experimental results with realistic music show the average accuracy of 76.2% for the instrument annotation and that the instrogram-based similarity measure represents the actual instrumentation similarity better than an MFCC-based one. ©2006 IEEE.

    DOI

  • ロボットによる周囲状況把握のための雑音下での環境音認識

    海尻聡, 石原一志, 北原鉄朗, Valin Jean-Marc, 駒谷和範, 尾形哲也, 奥乃博

    第6回システムインテグレーション部門講演会 (SI2005), 計測自動制御学会    2005年12月

  • 音源分離と音源認識を統合した同時発話認識システムの同時発話文による評価

    山本俊一, 中臺一博, 中野幹生, 辻野広司, Valin Jean-Marc, 駒谷和範, 尾形哲也, 奥乃博

    第6回システムインテグレーション部門講演会 (SI2005), 計測自動制御学会    2005年12月

  • 自己組織化論理回路における学習アルゴリズムに関する解析

    金天海, 出澤純一, 尾形哲也, 菅野重樹

    第6回システムインテグレーション部門講演会 (SI2005), 計測自動制御学会    2005年12月

  • 超音波センサアレイを用いたアクティブセンシングによる3次元物体の位置形状認識

    丹羽治彦, 駒谷和範, 尾形哲也, 奥乃博

    音響学会関西支部第8回若手研究者交流研究発表会    2005年12月

  • 多重奏音源同定における音の重なりに対する頑健性の改善

    北原鉄朗, 後藤真考, 駒谷和範, 尾形哲也, 奥乃博

    音響学会関西支部第8回若手研究者交流研究発表会    2005年12月

  • 非線形振動子による引き込みを利用した仮想空間における歩行

    小鷹研理, 尾形哲也, 奥乃博

    ヒューマンインタフェース学会論文誌   7 ( 4 ) 443 - 451  2005年11月

  • モダリティ間マッピングを用いた環境音を表現するロボットジェスチャー生成

    服部佑哉, 小嶋秀樹, 駒谷和範, 尾形哲也, 奥乃博

    第23回日本ロボット学会学術講演会    2005年09月

  • 人間との対話のためのロボットによる疑似シンボルの獲得とリアルタイム表情認識

    松本祥平, 大谷拓, 駒谷和範, 尾形哲也, 奥乃博

    第23回日本ロボット学会学術講演会    2005年09月

  • 身体の動力学的特性に基づいた運動プリミティブの獲得-2自由度リンク系のランダムな運動の分節化の試み

    有江浩明, 尾形哲也, 菅野重樹

    第23回日本ロボット学会学術講演会    2005年09月

  • ロボットによる複数人同時インタラクションのための人間新密度空間マッピング

    田崎豪, 駒谷和範, 尾形哲也, 奥乃博

    第23回日本ロボット学会学術講演会    2005年09月

  • RNNPBによるマルチモーダルダイナミクス表現に基づいたヒューマノイドの物体能動知覚

    尾形哲也, 大庭隼人, 谷淳, 駒谷和範, 奥乃博

    第23回日本ロボット学会学術講演会    2005年09月

  • 複数のカルマンフィルタを用いた複数移動話者追跡と精度評価

    村瀬昌満, 山本俊一, Valin Jean-Marc, 中臺一博, 山田健太郎, 駒谷和範, 尾形哲也, 奥乃博

    第23回日本ロボット学会学術講演会    2005年09月

  • 実ロボットを用いた対話型進化的計算によるコミュニケーション学習

    生熊良規, 長尾大輔, 菅佑樹, 尾形哲也, 菅野重樹

    第23回日本ロボット学会学術講演会    2005年09月

  • ミッシングフィーチャー理論に基づく音源分離と音声認識インタフェースの評価

    山本俊一, Valin Jean-Marc, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    第23回日本ロボット学会学術講演会    2005年09月

  • 混合音からの特徴量テンプレート作成と音楽的文脈利用による多重奏の音源同定

    北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    音響学会秋期研究発表会   3-10-15  2005年09月

  • 伴奏音抑制と高信頼度フレーム選択に基づく楽曲中の歌声の歌手名同定手法

    藤原弘将, 北原鉄朗, 後藤真考, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会, 音楽情報科学研究報告   2005-MUS-61-16, Vol. 2005 ( No. 82 )  2005年08月

  • ドラムパターン推定によるドラム音認識誤り補正手法

    吉井和佳, 後藤真考, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会, 音楽情報科学研究報告   2005-MUS-61-16, Vol. 2005 ( No. 82 )  2005年08月

  • Generating Confirmation to Distinguish Phonologically Confusing Word Pairs in Spoken Dialogue Systems

    Kazunori KOMATANI, Ryoji HAMABE, Tetsuya OGATA, Hiroshi G. OKUNO

    Proc. of 4th IJCAI Workshop on Knowledge and Reasoning in Practical Dialogue Systems     40 - 45  2005年07月

  • Robot Gesture Generation from Environmental Sounds Using Inter-modality Mapping

    Yuya HATTORI, Hideki KOZIMA, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proc. International Workshop on Epigenetic Robotics     139 - 140  2005年05月

  • 音声対話システムにおける音韻的類似表現の混同を防ぐための確認の自動生成

    浜辺良二, 駒谷和範, 尾形哲也, 奥乃博

    第56回音声言語情報処理研究会   2005-SLP-Vol. 2005  2005年05月

  • 環境音の擬音語変換のための環境音用音素の設計

    石原一志, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会論文誌   20 ( 3 ) 229 - 236  2005年03月

  • 音声対話システムにおける音韻的類似度の混同を防ぐための確認の生成

    浜辺良二, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • データベース検索タスクの文脈的制約を用いた音声対話システムの実験的評価

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • RNNによるロボットの物体操作のための物体ダイナミクスの抽出

    大庭隼人, 駒谷和範, 尾形哲也, 谷淳, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • ロボットとの協調作業のためのRNNによる疑似シンボルの獲得

    大谷拓, 駒谷和範, 尾形哲也, 谷淳, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 音声情報と顔画像情報の統合によるパラ言語レベルの感情認識

    松本祥平, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • ロボットによる人間新密度の空間マッピングとインタラクションへの適用

    田崎豪, 松本祥平, 大庭隼人, 村瀬昌満, 大谷拓, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 方向と距離の変化に対してロバストな唇認識

    山口健, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • マイクロフォンアレイを用いて分離した環境音による状況認識

    海尻聡, 石原一志, Jean-Marc Valin, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • カルマンフィルタによる音声の時系列特徴を用いた複数移動話者の追跡

    村瀬昌満, 山本俊一, Jean-Marc Valin, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • マイクロフォンアレイによる分離音認識のためのミッシングフィーチャーマスク自動生成

    山本俊一, Jean-Marc Valin, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • ドメイン間マッピングによる環境音に呼応したロボット動作生成

    服部佑哉, 小嶋秀樹, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 擬音語自動認識に基づいた環境音検索システム

    石原一志, 服部佑哉, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 簡易的インタフェースを用いた仮想空間移動のための歩行感覚の補償

    小鷹研理, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 歌声の調波構造抽出を用いた歌手名同定

    藤原弘将, 北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 調波構造の抑制によるドラムス発音時刻検出の頑健化

    吉井和佳, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 多重奏の音源同定のための混合音からのテンプレート作成法

    北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第67回全国大会    2005年03月

  • 超音波センサアレイを用いたアクティブセンシングによる3次元物体の位置形状認識

    丹羽治彦, 駒谷和範, 尾形哲也, 奥乃博

    音響学会春期講演会    2005年03月

  • 音楽音響信号による室内音場共振周波数のブラインンド推定

    吉岡拓也, 引地孝文, 三好正人, 駒谷和範, 尾形哲也, 奥乃博

    音響学会春期講演会    2005年03月

  • データベース検索タスクの文脈的制約を用いた音声対話システムの実験的評価

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    音声言語情報処理研究会(情報研報)    2005年02月

  • Sound-imitation word recognition for environmental sounds disambiguation in determining phonemes of sound-imitation words

    Kazushi Ishihara, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Transactions of the Japanese Society for Artificial Intelligence   20 ( 3 ) 229 - 236  2005年  [査読有り]

     概要を見る

    Environmental sounds are very helpful in understanding environmental situations and in telling the approach of danger, and sound-imitation words (sound-related onomatopoeia) are important expressions to inform such sounds in human communication, especially in Japanese language. In this paper, we design a method to recognize sound-imitation words (SIWs) for environmental sounds. Critical issues in recognizing SIW are how to divide an environmental sound into recognition units and how to resolve representation ambiguity of the sounds. To solve these problems, we designed three-stage procedure that transforms environmental sounds into sound-imitation words, and phoneme group expressions that can represent ambiguous sounds. The three-stage procedure is as follows: (1) a whole waveform is divided into some chunks, (2) the chunks are transformed into sound-imitation syllables by phoneme recognition, (3) a sound-imitation word is constructed from sound-imitation syllables according to the requirements of the Japanese language. Ambiguity problem is that an environmental sound is often recognized differently by different listeners even under the same situation. Phoneme group expressions are new phonemes for environmental sounds, and they can express multiple sound-imitation words by one word. We designed two sets of phoneme groups: "a set of basic phoneme group" and "a set of articulation-based phoneme group" to absorb the ambiguity. Based on subjective experiments, the set of basic phoneme groups proved more appropriate to represent environmental sounds than the articulation-based one or a set of normal Japaneses phonemes.

    DOI

  • Open-end human-robot interaction from the dynamical systems perspective: Mutual adaptation and incremental learning

    Tetsuya Ogata, Shigeki Sugano, Jun Tani

    Advanced Robotics   19 ( 6 ) 651 - 670  2005年  [査読有り]

     概要を見る

    In this paper, we experimentally investigated the open-end interaction generated by the mutual adaptation between humans and robot. Its essential characteristic, incremental learning, is examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We used the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN) for the robot control. Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Next, we used a 'consolidation-learning algorithm' as a model of the hippocampus in the brain. In this method, the RNN was trained by both new data and the rehearsal outputs of the RNN not to damage the contents of current memory. The proposed method enabled the robot to improve performance even when learning continued for a long time (openend). The dynamical systems analysis of RNNs supports these differences and also showed that the collaboration scheme was developed dynamically along with succeeding phase transitions. © VSP and Robotics Society of Japan 2005.

    DOI

  • Enhanced robot speech recognition based on microphone array source separation and missing feature theory

    Shun'ichi Yamamoto, Jean Marc Valin, Kazuhiro Nakadai, Jean Rouat, François Michaud, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Conference on Robotics and Automation   2005   1477 - 1482  2005年  [査読有り]

     概要を見る

    A humanoid robot under real-world environments usually hears mixtures of sounds, and thus three capabilities are essential for robot audition; sound source localization, separation, and recognition of separated sounds. While the first two are frequently addressed, the last one has not been studied so much. We present a system that gives a humanoid robot the ability to localize, separate and recognize simultaneous sound sources. A microphone array is used along with a real-time dedicated implementation of Geometric Source Separation (GSS) and a multi-channel post-filter that gives us a further reduction of interferences from other sources. An automatic speech recognizer (ASR) based on the Missing Feature Theory (MFT) recognizes separated sounds in real-time by generating missing feature masks automatically from the post-filtering step. The main advantage of this approach for humanoid robots resides in the fact that the ASR with a clean acoustic model can adapt the distortion of separated sound by consulting the post-filter feature masks. Recognition rates are presented for three simultaneous speakers located at 2m from the robot. Use of both the post-filter and the missing feature mask results in an average reduction in error rate of 42% (relative). ©2005 IEEE.

    DOI

  • Distance-based dynamic interaction of humanoid robot with multiple people

    Tsuyoshi Tasaki, Shohei Matsumoto, Hayato Ohba, Mitsuhiko Toda, Kazuhiro Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   3533 LNAI   111 - 120  2005年  [査読有り]

     概要を見る

    Research on human-robot interaction is getting an increasing amount of attention. Because almost all the research has dealt with communication between one robot and one person, quite little is known about communication between a robot and multiple people. We developed a method that enables robots to communicate with multiple people by selecting an interactive partner using criteria based on the concept of proxemics. In this method, a robot changes active sensory-motor modalities based on the interaction distance between itself and a person. Our method was implemented in a humanoid robot, SIG2, using a subsumption architecture. SIG2 has various sensory-motor modalities to interact with humans. A demonstration of SIG2 showed that the proposed method works well during interaction with multiple people. © Springer-Verlag Berlin Heidelberg 2005.

    DOI

  • Self-organizing algorithm for logic circuit based on local rules

    Chyon Hae Kim, Tetsuya Ogata, Shigeki Sugano

    IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM   2   1192 - 1197  2005年  [査読有り]

     概要を見る

    This study discusses a learning algorithm for autonomous robots that has five characteristics including autonomous exploration of effective output, low calculation costs, capability for multi-tasking, reusing past knowledge, and handling time series. We propose the use of self-organizing network elements (SONE) as a method for creating learning systems that provide these characteristics. Using this method, we created and evaluated a Self-Organizing Logic Circuit. The results of our experiments showed that this learning system met the requirements by being capable of creating a basic logic circuit, learning additional knowledge, controlling a simple robot in a simulation, and solving a maze problem. ©2005 IEEE.

  • Making a robot recognize three simultaneous sentences in real-time

    Shun'Ichi Yamamoto, Kazuhiro Nakadai, Jean Marc Valin, Jean Rouat, François Michaud, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     4040 - 4045  2005年  [査読有り]

     概要を見る

    A humanoid robot under real-world environments usually hears mixtures of sounds, and thus three capabilities are essential for robot audition; sound source localization, separation, and recognition of separated sounds. We have adopted the missing feature theory (MFT) for automatic recognition of separated speech, and developed the robot audition system. A microphone array is used along with a real-time dedicated implementation of Geometric Source Separation (GSS) and a multi-channel post-filter that gives us a further reduction of interferences from other sources. The automatic speech recognition based on MFT recognizes separated sounds by generating missing feature masks automatically from the post-filtering step. The main advantage of this approach for humanoid robots resides in the fact that the ASR with a clean acoustic model can adapt the distortion of separated sound by consulting the post-filter feature masks. In this paper, we used the improved Julius as an MFT-based automatic speech recognizer (ASR). The Julius is a real-time large vocabulary continuous speech recognition (LVCSR) system. We performed the experiment to evaluate our robot audition system. In this experiment, the system recognizes a sentence, not an isolated word. We showed the improvement in the system performance through three simultaneous speech recognition on the humanoid SIG2. © 2005 IEEE.

    DOI

  • Interactive evolution of human-robot communication in real world

    Yuki Suga, Yoshinori Ikuma, Daisuke Nagao, Shigeki Sugano, Tetsuya Ogata

    2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1438 - 1443  2005年  [査読有り]

     概要を見る

    This paper describes how to implement interactive evolutionary computation (IEC) into a human-robot communication system. IEC is an evolutionary computation (EC) in which the fitness function is performed by human assessors. We used IEC to configure the human-robot communication system. We have already simulated IEC's application. In this paper, we implemented IEC into a real robot. Since this experiment leads considerable burdens on both the robot and experimental subjects, we propose the human-machine hybrid evaluation (HMHE) to increase the diversity within the genetic pool without increasing the number of interactions. We used a communication robot, WAMOEBA-3 (Waseda Artificial Mind On Emotion BAse), which is appropriate for this experiment. In the experiment, human assessors interacted with WAMOEBA- 3 in various ways. The fitness values increased gradually, and assessors felt the robot learnt the motions they desired. Therefore, it was confirmed that the IEC is most suitable as the communication learning system. © 2005 IEEE.

    DOI

  • Spatially mapping of friendliness for human-robot interaction

    Tsuyoshi Tasaki, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     1277 - 1282  2005年  [査読有り]

     概要を見る

    It is important that robots interact with multiple people. However, most research has dealt with only interaction between one robot and one person and assumed that the distance between them does not change. This paper focuses on the spatial relationships between a robot and multiple people during interaction. Based on the distance between them, our robot selects appropriate functions to use. It does this using a method we developed for spatially mapping the "friendliness" of each space around the robot. The robot interacts with the highest friendliness spaces (people) selectively, thereby enabling interaction between the robot and multiple people. Our humanoid robot, SIG2 which the proposed method was implemented into, interacted with about 30 visitors, at the Kyoto University Museum. The results obtained using questionnaires after interaction showed that the actions of SIG2 were easy to understand even when it interacted with multiple people at the same time and that SIG2 behaved in a friendly manner. © 2005 IEEE.

    DOI

  • Extracting multi-modal dynamics of objects using RNNPB

    Tetsuya Ogata, Hayato Ohba, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS     160 - 165  2005年  [査読有り]

     概要を見る

    Dynamic features play an important role in recognizing objects that have similar static features in colors and or shapes. This paper focuses on active sensing that exploits dynamic feature of an object. An extended version of the robot, Robovie-IIs, moves an object by its arm to obtain its dynamic features. Its issue is how to extract symbols from various kinds of temporal states of the object. We use the recurrent neural network with parametric bias (RNNPB) that generates selforganized nodes in the parametric bias space. The RNNPB with 42 neurons was trained with the data of sounds, trajectories, and tactile sensors generated while the robot was moving/hitting an object with its own arm. The clusters of 20 kinds of objects were successfully self-organized. The experiments with unknown (not trained) objects demonstrated that our method configured them in the PB space appropriately, which proves its generalization capability. © 2005 IEEE.

    DOI

  • Multiple moving speaker tracking by microphone array on mobile robot

    Masamitsu Murase, Shunichi Yamamoto, Jean Marc Valin, Kazuhiro Nakadai, Kentaro Yamada, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    9th European Conference on Speech Communication and Technology     249 - 252  2005年

     概要を見る

    Real-world applications often require tracking multiple moving speakers for improving human-robot interactions and/or sound source separation. This paper presents multiple moving speaker tracking using an 8ch microphone array system installed on a mobile robot. This problem is difficult because the system does not assume that sound sources and/or the microphone array are fixed. Our solutions consist of two key ideas - time delay of arrival estimation, and multiple Kalman filters. The former localizes multiple sound sources based on beamforming in real time. Non-linear movements are tracked by using a set of Kalman filters with different history lengths in order to reduce errors in tracking multiple moving speakers under noisy and echoic environments. For quantitative evaluation of the tracking, motion references of sound sources and a mobile robot, called SIG2, were measured accurately by ultrasonic 3D tag sensors. As a result, we showed that the system tracked three simultaneous sound sources even when SIG2 moved in a room with large reverberation due to glass walls.

  • Contextual constraints based on dialogue models in database search task for spoken dialogue systems

    Kazunori Komatani, Naoyuki Kanda, Tetsuya Ogata, Hiroshi G. Okuno

    9th European Conference on Speech Communication and Technology     877 - 880  2005年

     概要を見る

    This paper describes the incorporation of contextual information into spoken dialogue systems in the database search task. Appropriate dialogue modeling is required to manage automatic speech recognition (ASR) errors using dialogue-level information. We define two dialogue models: a model for dialogue flow and a model of structured dialogue history. The model for dialogue flow assumes dialogues in the database search task consist of only two modes. In the structured dialogue history model, query conditions are maintained as a tree structure, taking into consideration their inputted order. The constraints derived from these models are integrated by using a decision tree learning, so that the system can determine a dialogue act of the utterance and whether each content word should be accepted or rejected, even when it contains ASR errors. The experimental result showed that our method could interpret content words better than conventional one without the contextual information. Furthermore, it was also shown that our method was domain-independent because it achieved equivalent accuracy in another domain with-out any more training.

  • Singer identification based on accompaniment sound reduction and reliable frame selection

    Hiromasa Fujihara, Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISMIR 2005 - 6th International Conference on Music Information Retrieval     329 - 336  2005年

     概要を見る

    This paper describes a method for automatic singer identification from polyphonic musical audio signals including sounds of various instruments. Because singing voices play an important role in musical pieces with a vocal part, the identification of singer names is useful for music information retrieval systems. The main problem in automatically identifying singers is the negative influences caused by accompaniment sounds. To solve this problem, we developed two methods, accompaniment sound reduction and reliable frame selection. The former method makes it possible to identify the singer of a singing voice after reducing accompaniment sounds. It first extracts harmonic components of the predominant melody from sound mixtures and then resynthesizes the melody by using a sinusoidal model driven by those components. The latter method then judges whether each frame of the obtained melody is reliable (i.e. little influenced by accompaniment sound) or not by using two Gaussian mixture models for vocal and non-vocal frames. It enables the singer identification using only reliable vocal portions of musical pieces. Experimental results with forty popular-music songs by ten singers showed that our method was able to reduce the influences of accompaniment sounds and achieved an accuracy of 95%, while the accuracy for a conventional method was 53%. © 2005 Queen Mary, University of London.

  • Instrument identification in polyphonic music: Feature weighting with mixed sounds, pitch-dependent timbre modeling, and use of musical context

    Tetsuro Kitahara, Masataka Goto, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    ISMIR 2005 - 6th International Conference on Music Information Retrieval     558 - 563  2005年

     概要を見る

    This paper addresses the problem of identifying musical instruments in polyphonic music. Musical instrument identification (MII) is an improtant task in music information retrieval because MII results make it possible to automatically retrieving certain types of music (e.g., piano sonata, string quartet). Only a few studies, however, have dealt with MII in polyphonic music. In MII in polyphonic music, there are three issues: feature variations caused by sound mixtures, the pitch dependency of timbres, and the use of musical context. For the first issue, templates of feature vectors representing timbres are extracted from not only isolated sounds but also sound mixtures. Because some features are not robust in the mixtures, features are weighted according to their robustness by using linear discriminant analysis. For the second issue, we use an F0-dependent multivariate normal distribution, which approximates the pitch dependency as a function of fundamental frequency. For the third issue, when the instrument of each note is identified, the a priori probablity of the note is calculated from the a posteriori probabilities of temporally neighboring notes. Experimental results showed that recognition rates were improved from 60.8% to 85.8% for trio music and from 65.5% to 91.1% for duo music. © 2005 Queen Mary, University of London.

  • Walking with body-sense in virtual space using the nonlinear oscillator

    Kenri Kodaka, Tetsuya Ogata, Hiroshi G. Okuno

    Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics   1   324 - 329  2005年  [査読有り]

     概要を見る

    This paper presents a novel construction of a locomotion system that compensates for walking-sense with simple interfaces using hands or fingers. The realization of moving with body-sense in virtual space has required high-quality system designs including walking cancellations, haptic feedback, and high-resolution displays. However, such approaches result in increasing costs of calculation and space, which obstruct the spread of VR technology. We therefore propose a new framework of locomotion systems with simple interfaces that give users "passivity and restraint", which are essential components of walking-sense. They are realized with mutual entrainment between a nonlinear oscillator and users' input. Two experiments were conducted for evaluation. The first one showed that our system gives users sense of distance based on a body standard and sense of rhythm with stable input. The second one demonstrated that the users of our system can experience a subjective body-sense and a sense of velocity. © 2005 IEEE.

    DOI

  • Acquisition of Motion Primitives of Robot in Human-Navigation Task -Towards Human Robot Interaction based on ”Quasi-Symbol

    Tetsuya OGATA, Shigeki SUGANO, Jun TANI

    人工知能学会論文誌   20 ( 3 ) 188 - 196  2005年

    DOI

  • Dynamic communication of humanoid robot with multiple people based on interaction distance

    Tsuyoshi Tasaki, Shohei Matsumoto, Hayato Ohba, Shunichi Yamamoto, Mitsuhiko Toda, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Transactions of the Japanese Society for Artificial Intelligence   20 ( 3 ) 209 - 219  2005年

     概要を見る

    Research on human-robot interaction is getting an increasing amount of attention. Since most research has dealt with communication between one robot and one person, quite few researchers have studied communication between a robot and multiple people. This paper presents a method that enables robots to communicate with multiple people using the "selection priority of the interactive partner" based on the concept of Proxemics. In this method, a robot changes active sensory-motor modalities based on the interaction distance between itself and a person. Our method was implemented into a humanoid robot, SIG2. SIG2 has various sensory-motor modalities to interact with humans. A demonstration of SIG2 showed that our method selected an appropriate interaction partner during interaction with multiple people.

    DOI

  • Extracting Multimodal Dynamics of Objects Using RNNPB.

    Tetsuya Ogata, Hayato Ohba, Jun Tani, Kazunori Komatani, Hiroshi G. Okuno

    J. Robotics Mechatronics   17 ( 6 ) 681 - 688  2005年

    DOI

  • ローカルルールに基づいたネットワーク素子の生成淘汰アルゴリズム

    金天海, 尾形哲也, 菅野重樹

    計測自動制御学会第5回システムインテグレーション部門学術講演会    2004年12月

  • 情緒交流ロボットWAMOEBA-3 &#8211; 人間-ロボット間コミュニケーション研究プラットフォームの開発

    有江浩明, 菅佑樹, 尾形哲也, 菅野重樹

    計測自動制御学会第5回システムインテグレーション部門学術講演会    2004年12月

  • Robot Motion Control using Listener”s Back-Channels and Head Gesture Information

    Tsuyoshi TASAKI, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proc. of The second International Workshop on Man-Machine Symbiotic Systems     385 - 391  2004年11月

  • Dynamic Communication between Multiple People and Robots Based on Interaction Distance

    Tsuyoshi TASAKI, Shohei MATSUMOTO, Hayato OHBA, Mitsuhiko TODA, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proc. of The second International Workshop on Man-Machine Symbiotic Systems     327 - 337  2004年11月

  • Acquisition of Motion Primitives of Robot in Human-Navigation Task: Towards Human-Robot Interaction Based on ”Quasi-Symbol

    Tetsuya OGATA, Shigeki SUGANO, Jun TANI

    Proc. of The second International Workshop on Man-Machine Symbiotic Systems     315 - 326  2004年11月

  • Development of Emotional Communication Robot, WAMOEBA-3

    Yuki SUGA, Tetsuya OGATA, Shigeki SUGANO

    Proc. of International Conference on Advanced Mechatronics (ICAM 2004)     413 - 418  2004年10月

  • Flexible assembly work cooperating system based on work state identifications by a self-organizing map

    Yasuhisa Hayakawa, Tetsuya Ogata, Shigeki Sugano

    IEEE/ASME Transactions on Mechatronics   9 ( 3 ) 520 - 528  2004年09月  [査読有り]

     概要を見る

    This paper presents a method of realizing flexible assembly work cooperation where the assembler is free to carry out the work, without constraints in the process. To realize such systems, there exists an issue of identifying work states during the assembly and to determine when and what kind of support is necessary. As an approach to solve such issues we took a self-organizing approach in constructing a work model, as an abstract model describing typical work states during the assembly. The necessity of support is judged by detecting uncommon work states occurring, and the type of support is determined by detecting the work state. Examples of work state identifications by the self-organized map are shown. We carried out experiments to evaluate the judgment of situational necessity of support and to verify the correct identification rate of typical work states. Finally a robotic support system was constructed that gives supports of autonomously holding and handing out assembly pieces by the judging of situational necessity of support. © 2004 IEEE.

    DOI

  • データベース検索タスクにおける文脈的制約を用いた音声対話システム

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    FIT-2004「情報科学技術レターズ」    2004年09月

  • ロボットでの利用を目的とした顔画像情報と音声情報の統合による感情認識

    松本祥平, 山口健, 駒谷和範, 尾形哲也, 奥乃博

    第22回日本ロボット学会学術講演会    2004年09月

  • ロボットと複数の人とのインタラクションにおける距離に基づくコミュニケーション

    田崎豪, 松本祥平, 大庭隼人, 戸田充彦, 駒谷和範, 尾形哲也, 奥乃博

    第22回日本ロボット学会学術講演会    2004年09月

  • ミッシングフィーチャー理論に基づく音源分離と音声認識のインターフェースの評価

    山本俊一, 中臺一博, 辻野広司, 駒谷和範, 尾形哲也, 奥乃博

    第22回日本ロボット学会学術講演会    2004年09月

  • 動作プリミティブを介した人間とロボットの協調

    尾形哲也, 菅野重樹, 谷淳

    第22回日本ロボット学会学術講演会    2004年09月

  • 一次反射音が移動の感覚に与える影響について

    小鷹研理, 駒谷和範, 尾形哲也, 奥乃博

    日本認知科学会第21回大会     R - 22  2004年07月

  • 環境音の擬音語変換のための環境音用音素の設計

    石原一志, 中谷智宏, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会全国大会(18回)     1E2 - 03  2004年06月

  • 多方向の唇画像を利用した音声認識

    山口健, 山本俊一, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会全国大会(18回)     1E2 - 02  2004年06月

  • データベース検索タスクにおける文脈的制約を用いた音声対話システム

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    第41回 言語音声理解と対話処理研究会, 人工知能学会    2004年06月

  • Interactive ECを用いた自律移動ロボットの進化に関する基礎研究

    菅佑樹, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2P2  2004年06月

  • 動作模倣による人間とロボットのインタラクション-手先軌道データを用いた動作予測システムの構築

    菅佑樹, 秋和祐介, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2A1  2004年06月

  • 音楽音響信号を対象とした和音進行の認識

    吉岡拓也, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    日本音響学会, MA 研究会    2004年06月

  • 音楽音響信号を対象とした和音変化時刻と和音名の同時認識

    吉岡拓也, 吉井和佳, 北原鉄朗, 櫻庭洋平, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • ロボットの挙動選択のための顔情報と音声情報を統合した感情判別

    松本祥平, 山口健, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • 連続環境音の繰り返し構造の認識

    服部佑哉, 石原一志, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • マルチモーダル情報による相槌の認識とロボット対話への応用

    田崎豪, 山口健, 戸田充彦, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • 音声対話システムにおける話題の構造を用いた効率的な対話管理,

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • ミッシングフィーチャー理論による三話者同時発話認識の向上

    山本俊一, 中臺一博, 辻野広司, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • マルチモーダル情報統合によるヒューマノイドロボットの挙動選択

    戸田充彦, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • 唇情報を利用した混合音声の分離

    山口健, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • 反射音が自己運動感覚に与える影響の考察

    小鷹研理, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • 環境音の擬音語変換における音素決定曖昧性の解消

    石原一志, 服部佑哉, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • 自動採譜における音色と定位と音楽知識を統合したパート形成

    櫻庭洋平, 尾形哲也, 奥乃博

    情報処理学会第66回全国大会    2004年03月

  • Open-end human robot interaction from the dynamical systems perspective: Mutual adaptation and incremental learning

    Tetsuya Ogata, Shigeki Sugano, Jun Tani

    Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)   3029   435 - 444  2004年  [査読有り]

     概要を見る

    This paper describes interactive learning between human subjects and robot using the dynamical systems approach. Our research concentrated on the navigation system of a humanoid robot and human subjects whose eyes were covered. We used the recurrent neural network (RNN) for the robot control. We used a "consolidation-learning algorithm" as a model of hippocampus in brain. In this method, the RNN was trained by both a new data and the rehearsal outputs of the RNN, not to damage the contents of current memory. The proposed method enabled the robot to improve the performance even when learning continued for a long time (open-end). The dynamical systems analysis of RNNs supports these differences.

    DOI

  • Computational Auditory Scene Analysis and Its Application to Robot Audition

    Hiroshi G. Okuno, Tetsuya Ogata, Kazunori Komatani, Kazuhiro Nakadai

    Proceedings - International Conference on Informatics Research for Development of Knowledge Society Infrastructure, ICKS 2004     73 - 80  2004年  [査読有り]

     概要を見る

    We are engaged in research on computational auditory scene analysis to attain sophisticated robot (computer) human interaction by recognizing auditory awareness. The objective of our research is the understanding of an arbitrary sound mixture including non-speech sounds and music as well as voiced speech, obtained by robot's ears (or microphones embedded in the robot). The main issues are sound source localization, separation, and recognition at signal processing levels, and signal-to-symbol transformation at the interface level to symbol processing levels. The latter is critical in developmental communication and we are developing an automatic onomatopoeia recognition system. This paper overviews our activities in robot audition, in particular, active direction-pass filter (ADPF) that separates sounds originating from a specific direction by integrating sound source localization and visual processing. ADPF is implemented on three kinds of robots and demonstrates separating and recognizing three simultaneous speeches with a pair of microphones.

    DOI

  • Automatic sound-imitation word recognition from environmental sounds focusing on ambiguity problem in determining phonemes

    Kazushi Ishihara, Tomohiro Nakatani, Tetsuya Ogata, Hiroshi G. Okuno

    Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)   3157   909 - 918  2004年  [査読有り]

     概要を見る

    Sound-imitation words (SIWs), or onomatopoeia, are important for computer human interactions and the automatic tagging of sound archives. The main problem in automatic SIW recognition is ambiguity in the determining phonemes, since different listener hears the same environmental sound as a different SIW even under the same situation. To solve this problem, we designed a set of new phonemes, called the basic phoneme-group set, to represent environmental sounds in addition to a set of the articulation-based phoneme-groups. Automatic SIW recognition based on Hidden Markov Model (HMM) with the basic phoneme-groups is allowed to generate plural SIWs in order to absorb ambiguities caused by listener- and situation-dependency. Listening experiments with seven subjects proved that automatic SIW recognition based on the basic phoneme-groups outperformed that based on the articulation-based phoneme-groups and that based on Japanese phonemes. The proposed system proved more adequate to use computer interactions. © Springer-Verlag Berlin Heidelberg 2004.

    DOI

  • Dynamic communication of humanoid robot with multiple people based on interaction distance

    Tsuyoshi Tasaki, Shohei Matsumoto, Hayato Ohba, Mitsuhiko Toda, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Workshop on Robot and Human Interactive Communication     71 - 76  2004年  [査読有り]

     概要を見る

    Research on human-robot interaction is getting an increasing amount of attention. Since almost all the research has dealt with only communication between one robot and one person, there have been quite few discussions about communication between a robot and multiple people. This paper proposes a method which enables robots to communicate with multiple people using the 'selection priority of the interactive partner' based on the concept of 'Proxemics'. In this method, a robot changes active sensory-motor modalities based on the 'interaction distance' information. The proposed method is implemented into a humanoid robot SIG2 using subsumption architecture. SIG2 has various sensory-motor modalities to interact with humans. A demonstration of SIG2 showed that the proposed method works well during interaction with multiple people. © 2004 IEEE.

  • Repeat recognition for environmental sounds

    Yuya Hattori, Kazushi Ishihara, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proceedings - IEEE International Workshop on Robot and Human Interactive Communication     83 - 88  2004年  [査読有り]

     概要を見る

    This paper focuses on recognition of repeats in continuous environmental sounds. Environmental sounds, that are very informative in our daily lives, often repeat several times. Repeat recognition for environmental sounds is essential for compact representation of those sounds and for forecasting the future. In our method, input environmental sound signal is partitioned into several units according to the shapes of the power envelopes, and the auditory distance between every pair of units is computed. Repeating parts are then detected by using the approximate matching algorithm. Experimental results showed the 73.3% of recognition rate in counting the repeats of 30 environmental sounds, the length of each of which ranged from 20 seconds to 2 minutes. © 2004 IEEE.

  • Imitation based human-robot interaction-roles of joint attention and motion prediction

    Yusuke Akiwa, Yuki Suga, Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Workshop on Robot and Human Interactive Communication     283 - 288  2004年  [査読有り]

     概要を見る

    Behavior imitation is crucial for the acquisition of intelligence as well as in communication. This paper describes two kinds of experiments of human-robot communication based on behavior imitation. One compared results obtained when the robot did and did not predict the experimental subject 's behaviors by using past datasets, and the other compared results obtained with and without target objects in the simulator environment. The result of former experiment showed that the prediction of the subject's behaviors increase the subject's interest. The result of the latter experiment confirmed that the presence of objects facilitates joint attention and make human-robot communication possible even when the robot uses a simple imitation mechanism. This result shows that in human-robot communication, human not only recognizes the behaviors of the robot passively but also adapts to the situation actively. In conclusion, it is confirmed that motion prediction and the presence of objects for joint attention are important for human-robot communication. © 2004 IEEE.

  • Acquisition of reactive motion for communication robots using interactive EC

    Yuki Suga, Tetsuya Ogata, Shigeki Sugano

    2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)   2   1198 - 1203  2004年

     概要を見る

    We've developed an emotional communication robot, WAMOEBA, using behavior-based techniques. We also proposed motor-agent (MA) model, which is an autonomous distributed-control algorithm constructed of simple sensor-motor coordination. Though it enables WAMOEBA to behave in various ways, the weight of the combinations between different motor agents is influenced by the preferences of the developer. We usually use machine-learning algorithms to automatically configure these parameters for communication robots. However, this makes it difficult to define the quantitative evaluation required for communication. We therefore used the method of interactive evolutionary computation (IEC), which can be applied to problems involving quantitative evaluation. IEC does not require to define a fitness function; this task is performed by users. But the biggest problem with using IEC is human fatigue, which causes insufficiency of individuals and generations for convergence of EC. To fix this problem, we use the prediction function that automatically calculates the fitness values of genes from some samples that have received the human subjective evaluation. Then we carried out the behavior acquisition experiment using the IEC simulation system with the prediction function. As the results of experiments, it is confirmed that diversifying the genetic pool is an efficient way for generating a variety of behavior.

    DOI

  • Human-robot communication using multiple recurrent neural networks

    Yoshihiro Sakamoto, Tetsuya Ogata, Shigeki Sugano

    2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)   2   1574 - 1579  2004年

     概要を見る

    On the methodology of robotic design from the traditional view of communication which is assumed as a symbol process, robots are forced to confront the symbol grounding problem. However, if communication is assumed as the analog dynamics and robots are driven by it, robots can avoid the problem and be situated in the environment and to other agents. In this paper we will introduce a new communication system constructed from the view of dynamical systems to achieve the situatedness. This system is that there is a robot in a virtual environment and the control of the robot is shared by human operation using a joystick and a robot controller. As the controller, we adopt multiple recurrent neural networks (MRNN) which are able to cope with complex environments and broad communication that single recurrent net cannot cope with. We conduct two experiments in order to evaluate the effectiveness of MRNN to a low level communication task such as nonverbal interaction. First, we examine the effect of the number of RNNs contained in MRNN. Second, we examine the effect of the context dependency of MRNN. These experiments show the capability of MRNN as a new-type controller of communication robot.

    DOI

  • Human-robot collaboration using behavioral primitives

    Tetsuya Ogata, Masaki Matsunaga, Shigeki Sugano, Jun Tani

    2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)   2   1592 - 1597  2004年

     概要を見る

    A novel approach to human-robot collaboration based on quasi-symbolic expressions is proposed. The target task is navigation in which a person with his or her covered and a humanoid robot collaborate in a context-dependent manner. The robot uses a recurrent neural net with parametric bias (RNNPB) model to acquire the behavioral primitives, which are sensory-motor units, composing the whole task. The robot expresses the PB dynamics as primitives using symbolic sounds, and the person influences these dynamics through tactile sensors attached to the robot Experiments with six participants demonstrated that the level of influence the person has on the PB dynamics is strongly related to task performance, the person's subjective impressions, and the prediction error of the RNNPB model (task stability). Simulation experiments demonstrated that the subjective impressions of the correspondence between the utterance sounds (the PB values) and the motions were well reproduced by the rehearsal of the RNNPB model.

    DOI

  • Disambiguation in determining phonemes of sound-imitation words for environmental sound recognition

    Kazushi Ishihara, Yuya Hattori, Tomohiro Nakatani, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    8th International Conference on Spoken Language Processing, ICSLP 2004     1485 - 1488  2004年

     概要を見る

    Onomatopoeia, or sound-imitation words (SIWs) are important in informing sound events in human-computer communication. One problem is listener-dependency in recognizing environmental sounds by means of SIWs, that is, different listener hears the same environmental sound as a different SIW even under the same condition. Therefore, the use of usual Japanese phonemes is not adequate to express SIWs. To cope with this ambiguity problem of phoneme determination, we designed a set of new phonemes, referred to as the basic phoneme-groups, to represent environmental sounds. The basic phoneme-group consists of one or more Japanese phonemes, and thus the ambiguity problem is resolved based on it by generating one or more SIWs for a sound event. An HMM-based scheme is adopted to recognize SIWs using the phoneme-groups. Listening experiments with seven subjects showed that automatic SIW recognition based on the basic phoneme-groups outperformed ones based on the other types of phonemes. The recall and precision rate were 56.4% and 72.2%, respectively.

  • Robot Motion Control using Listener”s Back-Channels and Head Gesture Information

    Tsuyoshi TASAKI, Takeshi YAMAGUCHI, Kazunori KOMATANI, Tetsuya OGATA, Hiroshi G. OKUNO

    Proc. of International Conference on Spoken Language Processing (ICSLP-2004)     1033 - 1036  2004年

  • Automatic Chord Transcription with Concurrent Recognition of Chord Symbols and Boundaries.

    Takuya Yoshioka, Tetsuro Kitahara, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

    Proc. of 2004 International Conference on Musical Information Retrieval (ISMIR-2004)     100 - 105  2004年

  • Constructivist approach to human-robot emotional communication - Design of evolutionary function for WAMOEBA-3

    Yuki Suga, Hiroaki Arie, Tetsuya Ogata, Shigeki Sugano

    2004 4th IEEE-RAS International Conference on Humanoid Robots   2   869 - 884  2004年  [査読有り]

     概要を見る

    By applying a self-preservation function and communication capability, we investigated the emergence of the emotional behavior of robots to propose an "evaluation function for self-preservation," a "model of endocrine system," and a "MA model." We also developed a, new hardware platform, WAMOEBA-3, to install our new knowledge into the robot. WAMOEBA-3, a wheel type, independent robot, was designed for easy maintenance and customization. As a new function for WAMOEBA-3, we introduced an evolutionary function for the acquisition of reactive motion that uses an interactive evolutionary computation method. We show the results of simulation experiments and discuss real world applications. © 2004 IEEE.

    DOI

  • Computational Auditory Scene Analysis and Its Application to Robot Audition

    Hiroshi G. Okuno, Tetsuya Ogata, Kazunori Komatani, Kazuhiro Nakadai

    Proceedings - International Conference on Informatics Research for Development of Knowledge Society Infrastructure, ICKS 2004     73 - 80  2004年

     概要を見る

    We are engaged in research on computational auditory scene analysis to attain sophisticated robot (computer) human interaction by recognizing auditory awareness. The objective of our research is the understanding of an arbitrary sound mixture including non-speech sounds and music as well as voiced speech, obtained by robot's ears (or microphones embedded in the robot). The main issues are sound source localization, separation, and recognition at signal processing levels, and signal-to-symbol transformation at the interface level to symbol processing levels. The latter is critical in developmental communication and we are developing an automatic onomatopoeia recognition system. This paper overviews our activities in robot audition, in particular, active direction-pass filter (ADPF) that separates sounds originating from a specific direction by integrating sound source localization and visual processing. ADPF is implemented on three kinds of robots and demonstrates separating and recognizing three simultaneous speeches with a pair of microphones.

    DOI

  • コミュニケーションを指向した反射的行動の獲得-対話型進化計算を用いたアプローチの提案

    菅佑樹, 尾形哲也, 菅野重樹

    計測自動制御学会第4回システムインテグレーション部門学術講演会     1004 - 1005  2003年12月

  • 対話型進化的計算手法を用いたロボットの反射行動の獲得

    菅佑樹, 尾形哲也, 菅野重樹

    第21回日本ロボット学会学術講演会     3J31  2003年09月

  • 人間とロボットのインタラクティブな学習に関する研究

    尾形哲也, 真砂紀孝, 菅野重樹

    第21回日本ロボット学会学術講演会     1I26  2003年09月

  • Collaboration Development through Interactive Learning between Human and Robot

    Tetsuya OGATA, Noritaka MASAGO, Shigeki SUGANO, Jun TANI

    Proc. of 3rd International Workshop on Epigenetic Robotics     99 - 106  2003年08月

  • 情緒交流ロボットWAMOEBA-3の開発?トルクセンサ内蔵関節の設計

    土屋尚文, 有江浩明, 日野貴司, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2P2  2003年05月

  • メカノクリーチャー -生物から学ぶデザインテクノロジー-

    尾形哲也, 菅野重樹

    日本機会学会     172 - 203  2003年

  • Robust modeling of dynamic environment based on robot embodiment

    Kuniaki Noda, Mototaka Suzuki, Naofumi Tsuchiya, Yuki Suga, Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Conference on Robotics and Automation   3   3565 - 3570  2003年  [査読有り]

     概要を見る

    Recent studies on embodied cognitive science have shown us the possibility of emergence of more complex and nontrivial behaviors with quite simple designs if the designer takes the dynamics of the system-environment interaction into account properly. In this paper, we report our tentative classification experiments of several objects using the human-like autonomous robot, "WAMOEBA-2Ri". As modeling the environment, we focus on not only static aspects of the environment but also dynamic aspects of it including that of the system own. The visualized results of this experiment shows the integration of multimodal sensor dataset acquired by the system-environment interaction ("grasping") enable robust categorization of several objects. Finally, in discussion, we demonstrate a possible application to making "invariance in motion" emerge consequently by extending this approach.

    DOI

  • Flexible assembly work cooperation based on work state identifications by a self-organizing map

    Y. Hayakawa, T. Ogata, S. Sugano

    IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM   2   1031 - 1036  2003年  [査読有り]

     概要を見る

    This study presents a method of realizing flexible assembly work cooperation in cases where neither the assembly process, nor the final form of the completed task is pre-defined in advance. To realize such systems, there exists an issue of identifying work states during the assembly, and determine when and what kind of support is necessary. As an approach of solving such issues, identifying the work states from a work model built by self-organizing assembly motions of human is taken. Examples of the work state identifications and a support system, which judges the situational necessity of support and selects whether to hand out or holds assembly parts are shown. The support is carried out based on the work states identified by the self-organized map. Experiments indicate that work state identification by a self-organizing map is effective in flexibly cooperating with human during assembly work.

    DOI

  • Interactive Learning in Human-Robot Collaboration

    Tetsuya Ogata, Noritaka Masago, Shigeki Sugano, Jun Tani

    IEEE International Conference on Intelligent Robots and Systems   1   162 - 167  2003年  [査読有り]

     概要を見る

    In this paper, we investigated interactive learning between human subjects and robot experimentally, and its essential characteristics are examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We compared the usual feed-forward neural net-work (FFNN) without recursive connections and the recurrent neural network (RNN). Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental impressions, especially from the aspect of operability. When the robot used a consolidation-learning algorithm using the rehearsal outputs of the RNN, the performance improved even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject's mental impressions of the RNN improved significantly. The dynamical systems analysis of RNNs support these differences.

    DOI

  • 身体性に基づいた環境・ロボット自身における新奇性検出

    野田邦昭, 鈴木基高, 尾形哲也, 菅野重樹

    第20回日本ロボット学会学術講演会,     1C31  2002年10月

  • 模倣を主体とした人間とロボットのコミュニケーション?コミュニケーションにおける共同注意の役割

    大竹正海, 坂本義弘, 尾形哲也, 菅野重樹

    日本機械学会 ロボティクスメカトロニクス講演会     1P1 - K03  2002年06月

  • 分散エージェントを用いた全身協調による動作生成

    小宮孝章, 野田邦昭, 土屋尚文, 尾形哲也, 菅野重樹

    日本機械学会 ロボティクスメカトロニクス講演会     2P1 - D06  2002年06月

  • Development of Passive Elements with Variable Mechanical Impedance for Wearable Robots.

    S. Kawamura, T. Yamamoto, D. Ishida, Tetsuya Ogata, Y. Nakayama, Osamu Tabata, Susumu Sugiyama

        248 - 253  2002年

    DOI

  • Influence of the Eye Motions in Human-Robot Communication and Motion Generation based on the Robot Body Structure

    Tetsuya OGATA, Takaaki KOMIYA, Kuniaki NODA, Shigeki SUGANO

    Proc. of IEEE/RAS International Conference on Humanoid Robots (Humanoid 2001)     83 - 89  2001年11月

  • 模倣を主体とした人間─ロボットコミュニケーションに関する基礎実験

    大竹正海, 山本大介, 尾形哲也, 菅野重樹

    第19回日本ロボット学会学術講演会     109 - 110  2001年09月

  • 内分泌系モデルによる強化学習パラメータの動的調整

    尾形哲也, 松本典剛, 菅野重樹

    第19回日本ロボット学会学術講演会     1239 - 1240  2001年09月

  • 身体性に基づいた状態表現機能を持つロボットと人間とのコミュニケーション

    野田邦昭, 井田真高, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     1P1 - D10  2001年06月

  • 感性情報の動的変化に対応したモーションプラニングの実現と評価

    尾形哲也, 志村明俊, 渋谷恒司, 菅野重樹

    第6回ロボティクスシンポジア     402 - 407  2001年03月

  • Motion generation of the autonomous robot based on body structure

    Tetsuya Ogata, Takaaki Komiya, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems   4   2338 - 2343  2001年  [査読有り]

     概要を見る

    This study aims to investigate the intelligence which can make robots adapt to the human environment. This paper points out the problems of the behavior-based robot, and proposes the methods which can generate whole body motions based on body structure and integrates the reflection motions to make the behaviors continuous. The motion performances are compared in two kinds of environments, such as a dynamic environment and a static environment, by using a simulator of the autonomous robot WAMOEBA-2Ri developed in this research. Finally, we show that the integration parameters of the proposal method reflect the body structure of the robot and environmental structures.

    DOI

  • 人間とロボットの情緒的コミュニケーションの実験的評価?アームハンドによる人間との物理的インタラクション

    尾形哲也, 菅野重樹

    システム制御情報学会論文誌   13 ( 12 ) 566 - 474  2000年12月

  • 情動表現生成のためのロボット構造とシステムの統合

    尾形哲也, 菅野重樹

    計測自動制御学会SIシンポジウム     35 - 36  2000年12月

  • フェーズの 印象変化を考慮したバイオリン演奏アルゴリズムの構築

    尾形哲也, 志村明俊, 宇野格, 渋谷恒司, 菅野重樹

    第21回バイオメカニズム学会学術講演会     173 - 176  2000年11月

  • 情緒交流と自律行動生成のためのロボット構造

    尾形哲也, 菅野重樹

    第6回 IFTMM会議シンポジウム     14 - 17  2000年11月

  • Emotional communication between humans and the autonomous robot W AMOEBA-2 (Waseda amoeba) which has the emotion model

    Tetsuya Ogata, Shigeki Sugano

    JSME International Journal, Series C: Mechanical Systems, Machine Elements and Manufacturing   43 ( 3 ) 568 - 574  2000年09月  [査読有り]

     概要を見る

    In this paper, we discuss the communication between autonomous robots and humans through the development of a robot which has an emotion model. The model depicts the endocrine system of humans and has four kinds of hormone parameters to adjust various internal conditions such as motor output, cooling fan output and sensor gain. We surveyed 126 visitors at the '97 International Robot Exhibition held in Tokyo, Japan (Oct. 1997) to evaluate their psychological impressions of the robot. As a result, the human friendliness of the robot was confirmed and some factors of human-robot emotional communication were discovered.

    DOI

  • Emotional Communication Robot: WAMOEBA-2R &#8211; Emotion Model and Evaluation Experiments-

    Tetsuya OGATA, Shigeki SUGANO

    Proc. of IEEE/RAS International Conference on Humanoid Robots (Humanoid 2000)   93  2000年09月

  • 人間と自律ロボットのコミュニケーションに関する実験的考察?システム設計と心理評価の異母集団比較

    尾形哲也, 松山佳彦, 小宮孝章, 井田真高, 野田邦昭, 菅野重樹

    第18回日本ロボット学会学術講演会     479 - 480  2000年09月

  • 身体性に基づくロボットと人間とのコミュニケーション

    尾形哲也, 菅野重樹

    bit別冊「身体性とコンピュータ」,岡田美智男,三嶋博之,佐々木正人編,共立出版     195 - 207  2000年08月

  • The adaptive motion by the endocrine system model in an autonomous robot

    Tetsuya OGATA, Shigeki SUGANO

    Proc of International Symposium on Adaptive Motion of Animals and Machines     E30  2000年08月

  • 自律ロボット WAMOEBA-2Rの開発?アームシステムの搭載と心理実験

    尾形哲也, 松山佳彦, 小宮孝章, 井田真高, 野田邦昭, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     1A1  2000年05月

  • 未知作業支援に適用可能な作業状態モデル生成手法に関する研究

    第5回ロボティクスシンポジア   pp.260-265  2000年03月

  • 情緒交流ロボットWamoeba-2Rの開発〜システム構成と評価実験

    第5回ロボティクスシンポジア   pp.68-73  2000年03月

  • 情緒交流ロボットWamoeba-2Rの開発?システム構成と評価実験

    尾形哲也, 松山佳彦, 小宮孝章, 井田真高, 野田邦昭, 菅野重樹

    第5回ロボティクスシンポジア   pp.68-73   68 - 73  2000年03月

  • 未知作業支援に適用可能な作業状態モデル生成手法に関する研究

    早川泰久, 佐竹賢亮, 内田信宏, 尾形哲也, 菅野重樹

    第5回ロボティクスシンポジア   pp.260-265   260 - 265  2000年03月

  • Violin playing algorithm considering the change of phrase impression

    Tetsuya Ogata, Akitoshi Shimura, Koji Shibuya, Shigeki Sugano

    Proceedings of the IEEE International Conference on Systems, Man and Cybernetics   2   1342 - 1347  2000年

     概要を見る

    This study focused on the dynamics of KANSEI information and aims to propose the algorithm of motion planning using KANSEI. Concretely, the violin playing is regarded as the target motion which will be greatly influenced by KANSEI. This study introduced the multi-agent algorithm in which four physical bowing parameters are agents to adapt the impression transition smoothly maintaining the relationships between the parameters. We realized the violin performance suiting for the timbre words by introducing the proposed agent algorithm into the bowing machine developed in this research. As a result of the experiments, it was confirmed that there were various playing performances according to a single impression transition.

  • Assembly Support Based on Human Model -Provision of Physical Support According to Implicit Desire for Support-.

    Yasuhisa Hayakawa, Ikuo Kitagishi, Yusuke Kira, Kensuke Satake, Tetsuya Ogata, Shigeki Sugano

    J. Robotics Mechatronics   12 ( 2 ) 118 - 125  2000年

    DOI

  • Robotic co-operation system based on a self-organization approached human work model

    Yasuhisa Hayakawa, Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Conference on Robotics and Automation   4   4057 - 4062  2000年

     概要を見る

    This study presents a method of human co-operating systems, which can determine when support behavior is necessary by a human work model. We focus on assembly work as the target and propose a self-organizing approach of human work models by sampled human information by vision sensors. The support is determined according to the work model. Such a system would realize provision of support without a strict model of the assembly target and enable support in cases where neither the assembly process, nor the final form of the completed task, is known to the system in advance. First, a method of measuring human information and extracting states where support is necessary, from the human work model is presented. Next, a support system for assembly work co-operation according to the work model, with physical interaction capabilities is described. Experiments were carried out to evaluate and verify the effectiveness of the system. The results show that the constructed assembly support system is effective in both improving performance and increasing friendliness.

    DOI

  • Acquisition of internal representation in robots - toward human-robot communication using primitive language

    Tetsuya Ogata, Yoshihiro Matsuyama, Shigeki Sugano

    Advanced Robotics   14 ( 4 ) 277 - 291  2000年  [査読有り]

     概要を見る

    This research aims to clarify behavior intelligence and human cooperation intelligence of robots by emotion models which are based on the robot's hardware structure. In this paper, a human's mental image (internal expression) is given consideration as a method for emotional expression of robots. The hypothesis model for the acquisition of the internal expression of robots and experimental results using a real autonomous robot are described.

    DOI

  • A violin playing algorithm considering the change of phrase impression.

    Tetsuya Ogata, Akitoshi Shimura, Koji Shibuya, Shigeki Sugano

    SMC 2000 CONFERENCE PROCEEDINGS: 2000 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN & CYBERNETICS, VOL 1-5     1342 - 1347  2000年  [査読有り]

     概要を見る

    This study focused on the dynamics of KANSEI information and aims to propose the algorithm of motion planning using KANSEI Concretely: the violin playing is regarded as the target motion which will be greatly influenced by KANSEI. This study introduced the multi-agent algorithm in which four physical bowing parameters are agents to adapt the impression transition smoothly maintaining the relationships between the parameters. Ire realized the violin performance suiting for the timbre words by introducing the proposed agent algorithm into the bowing machine developed in this research. As a result of the experiments, it was confirmed that there were various playing performances according to a single impression transition.

    DOI

  • Development of arm system for human-robot emotional communication

    T. Ogata, T. Komiya, S. Sugano

    IECON Proceedings (Industrial Electronics Conference)   1   475 - 480  2000年  [査読有り]

     概要を見る

    This study aims to clarify the cooperation intelligence of robots with an emotion model based on the human biological system. The paper describes the functions of the arm system of the behavior-based robot, WAMOEBA-2, which has an emotion model and can communicate with humans. The specifications of the arm system were determined based on a study of human robot physical interaction. Each joint of the arm is equipped with a torque sensor and the arm is controlled by a distributed agent network system. The network architecture is acquired in a neural network by feedback-error-learning algorithm. In the experiment, human beings can play with the arm system by physical interaction.

    DOI

  • Development of emotional communication robot: WAMOEBA-2R Experimental evaluation of the emotional communication between robots and humans

    T. Ogata, Y. Matsuyama, T. Komiya, M. Ida, K. Noda, S. Sugano

    IEEE International Conference on Intelligent Robots and Systems   1   175 - 180  2000年  [査読有り]

     概要を見る

    This study aims to clarify the cooperation intelligence of robots. This paper describes the autonomous robot named WAMOEBA-2R which can communicate with humans by both an informational and physical way. WAMOEBA-2R has two arms of which each joint has a torque sensor to realize the physical interaction with humans, the function of the voice recognition and the face recognition. The arms are controlled by a distributed agent network system. The network architecture is acquired in a neural network by the feedback-error-learning algorithm. We surveyed 150 visitors at the '99 International Robot Exhibition held in Tokyo (Oct. 1999) to evaluate their psychological impressions of WAMOEBA-2R. As the result, some factors of the human-robot emotional communication were discovered.

    DOI

  • Analysis of Design Process by the Observation of Human Motion

    International Workshop on Emergent Synthesis (IWES'99)   pp.167-172  1999年12月

  • Analysis of Design Process by the Observation of Human Motion

    Tetsuya OGATA, Yasuhisa HAYAKAWA, Kensuke SATAKE, Shigeki SUGANO

    Proc. of International Workshop on Emergent Synthesis (IWES”99)     167 - 172  1999年12月

  • 人間-ロボット相互関係分析に基づくマニピュレータ支援行動生成

    第20回バイオメカニズム学会学術講演会   pp.380-381  1999年11月

  • 人間-ロボット相互関係分析に基づくマニピュレータ支援行動生成

    早川泰久, 内田信宏, 佐竹賢亮, 尾形哲也, 菅野重樹

    第20回バイオメカニズム学会学術講演会   pp.380-381   380 - 381  1999年11月

  • Extraction of Human Intention for Human Co-operating Systems -Prototype Assembling Work Support Robot System according to Human Intention

    International Conference on Advanced Robotics (ICAR'99)   pp. 199-204  1999年10月

  • Emotional Communication Between Humans and Robots - Consideration of Primitivee Language in Robots

    IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'99)   pp. 870-875  1999年10月

  • Human Robot Communication by Physical Interaction &#8211; Distributed Agent Control System and The Learning Algorithm -

    Tetsuya OGATA, Takaaki KOMIYA, Shigeki SUGANO

    Proc. of IEEE International Conference on Systems, Man, and Cybernetics (SMC”99)     1005 - 1010  1999年10月

  • Extraction of Human Intention for Human Co-operating Systems -Prototype Assembling Work Support Robot System according to Human Intention-

    Yasuhisa HAYAKAWA, Yusuke KIRA, Tetsuya OGATA, Shigeki SUGANO

    Proc. of International Conference on Advanced Robotics (ICAR”99)     199 - 204  1999年10月

  • 身体運動観察に基づく組立作業過程の記号化と分析

    第17回日本ロボット学会学術講演会   pp.1067-1068  1999年09月

  • 人間とロボットの物理的コミュニケーショ ン〜アームシステムの開発と分散エージェントによる行動獲得

    第17回日本ロボット学会学術講演会   pp. 423-424  1999年09月

  • 人間とロボットの物理的コミュニケーショ ン〜アームシステムの開発と分散エージェントによる行動獲得

    尾形哲也, 小宮孝章, 菅野重樹

    第17回日本ロボット学会学術講演会   pp. 423-424   423 - 424  1999年09月

  • 身体運動観察に基づく組立作業過程の記号化と分析

    早川泰久, 佐竹賢亮, 尾形哲也, 菅野重樹

    ,第17回日本ロボット学会学術講演会   pp.1067-1068   1067 - 1068  1999年09月

  • 情緒交流ロボットWamoeba-2のアームハンド機構の開発〜分散エージェントによる動作生成

    尾形哲也, 日塔潔, 小宮孝章, 松本典剛, 菅野重樹

    日本機械学会ロ ボティクスメカトロニクス講演会   2P2-79-103   2P2  1999年06月

  • 頭部・ 手部位置および把持情報の構造化による暗黙意図の抽出

    日本機械学会ロ ボティクスメカトロニクス講演会   2A1-76-118  1999年06月

  • 頭部・ 手部位置および把持情報の構造化による暗黙意図の抽出

    早川泰久, 吉良雄介, 佐竹賢亮, 内田信宏, 尾形哲也, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会     2A1  1999年06月

  • 情緒交流ロボット Wamoeba-2のアームハンド機構の開発?分散エージェントによる動作生成?

    尾形哲也, 日塔潔, 小宮孝章, 松本典剛, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会   2P2-79-103   2P2  1999年06月

  • 内分泌系モデルを組込んだロボットハードウェア

    尾形哲也, 菅野重樹

    バイオメカニズム学会誌   23 ( 2 ) 106 - 111  1999年05月

  • 情動モデルを有する自律ロボットWAMOEBA-2と人間との情緒交流

    尾形哲也, 菅野重樹

    日本機械学会論文誌C編   65 ( 633 ) 1900 - 1906  1999年05月

  • ロボットと人間の情緒交流—ロボットにおける前言語発生の考察

    第4回ロボティクス・シンポジア   PP44-49  1999年03月

  • 組立作業支援を目的としたシンセシスモデル構築に関する研究—組立動作情報を基にした時系列NNによる構造化

    早川泰久, 吉良雄介, 尾形哲也, 菅野重樹

    精密工学会春季大会学術講演会講演論文集   PP534   534  1999年03月

  • 組立作業支援を目的としたシ ンセシスモデル構築に関する研究?組立動作情報を基にした時系列NNによる構造化?

    早川泰久, 吉良雄介, 尾形哲也, 菅野重樹

    精密工学会春季大会学術講演会講演論文集   PP534   534  1999年03月

  • ロボットと人間の情緒交流—ロボットにおける前言語発生の考察

    尾形哲也, 松山佳彦, 菅野重樹

    第4回ロボティクス・シンポジア   PP44-49   44 - 49  1999年03月

  • 情動モデルを有する自律ロボットWAMOEBA−2と人間との情緒交流

    尾形哲也, 菅野重樹

    日本機械学会論文誌C編   Vol.65,No.633 ( 1900 ) 1906  1999年

    DOI

  • Emotional Communication between Humans and the Autonomous Robot WAMOEBA-2 (Waseda Amoeba) which has the Emotion Model

    Tetsuya Ogata, Shigeki Sugano

    Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C   65 ( 633 ) 1900 - 1906  1999年

     概要を見る

    This study discusses the communication between autonomous robots and humans through the development of a robot which has an emotion model. The model refers to the internal secretion system of humans and it has four kinds of the hormone parameters to use to adjust various internal conditions such as motor output, cooling fan output and sensor gain. We surveyed 126 visitors at ’97 International Robot Exhibition held in Tokyo (Oct. 1997) in order to exaluate psychological impressions of the robot. As the result, the human friendliness of the robot was confirmed and some factors concerning with the human-robot emotional communication were discovered. © 1999, The Japan Society of Mechanical Engineers. All rights reserved.

    DOI

  • Emotional communication between humans and the autonomous robot which has the emotion model

    Tetsuya Ogata, Shigeki Sugano

    Proceedings - IEEE International Conference on Robotics and Automation   4   3177 - 3182  1999年  [査読有り]

     概要を見る

    This study discusses the communication between autonomous robots and humans through the development of a robot which has an emotion model. The model refers to the internal secretion system of humans and it has four kinds of the hormone parameters to use to adjust various internal conditions such as motor output, cooling fan output and sensor gain. We surveyed 126 visitors at '97 International Robot Exhibition held in Tokyo, Japan (Oct. 1997) in order to evaluate psychological impressions of the robot. As a result, the human friendliness of the robot was confirmed and some factors of the human-robot emotional communication were discovered.

    DOI

  • Human-robot communication by physical interaction - distributed agent control system and the learning algorithm

    Tetsuya Ogata, Takaaki Komiya, Shigeki Sugano

    Proceedings of the IEEE International Conference on Systems, Man and Cybernetics   2  1999年

     概要を見る

    This study aims to clarify the cooperation intelligence of robots with the emotion model referring to the human biological system. This paper describes the functions of the arm system of the behavior-based robot, Wamoeba-2, which has an emotion model and can communicate with human. The specifications of the arm system were designed based on the consideration of the human robot physical interaction. Each joint of the arm is equipped with a torque sensor, and the arm is controlled by a distributed agent network system. The network architecture is acquired in a neural network by the feedback-error-learning algorithm. In the experiment, human beings can play with the arm system by physical interaction.

  • Emotional communication between humans and robots - consideration of primitive language in robots

    Tetsuya Ogata, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems   2   870 - 875  1999年

     概要を見る

    This research aims to clarify the behavior intelligence and the human cooperation intelligence of robots by the emotion models which is based on the robot's hardware structure. In this paper, human's mental images and language are given consideration as a method for emotional expression. The hypothesis model for the acquisition of the internal expressions of robots and the experimental results using a real autonomous robot are described.

    DOI

  • 人間のシンセシス活動の抽出とモデル化—人間機械協調システムを用いた一考察

    日本学術振興会未来開拓事業人間指向シンセシスの科学早稲田慶応成果報告会(第2回)講演論文集   PP70-73  1998年12月

  • 人間のシンセシス活動の抽出とモデル化?人間機械協調システムを用いた一考察?

    尾形哲也, 早川泰久, 吉良祐介, 菅野重樹

    日本学術振興会未来開拓事業 人間指向シンセシスの科学 早稲田慶応成果報告会(第2回) 講演論文集     70 - 73  1998年12月

  • ロボットにおけるホルモン系モデルとその影響—身体制御と環境適応

    バイオメカニズム学会学術講演会   PP71-72  1998年11月

  • 組立作業支援システムの構造—人間指向シンセシスの解釈

    日本機械学会設計工学・システム部門講演会   PP169-170  1998年11月

  • 組立作業支援システムの構造—人間指向シンセシスの解釈

    尾形哲也, 早川泰久, 菅野重樹

    日本機械学会設計工学・システム部門講演会   PP169-170   169 - 170  1998年11月

  • ロボットにおけるホルモン系モデルとその影響—身体制御と環境適応

    尾形哲也, 菅野重樹

    バイオメカニズム学会学術講演会   PP71-72   71 - 72  1998年11月

  • Emergence of Primitive Verbal Communication in Robots

    5th International Conference on Soft Computing and Information / Intelligent Systems   PP284-287  1998年10月

  • Mechanisums of Internal Secretion System for Intellectual Robots -Towards an Emergence of Emotion in Robots

    IEEE International Workshop on Robot and Human Communication   PP50-55  1998年10月

  • Mechanisms of Internal Secretion System for Intellectual Robots -Towards an Emergence of Emotion in Robots

    Tetsuya OGATA, Shigeki SUGANO

    Proc. of IEEE International Workshop on Robot and Human Communication (ROMAN”98)   PP50-55   50 - 55  1998年10月

  • Emergence of Primitive Verbal Communication in Robots

    Tetsuya OGATA, Shigeki SUGANO

    Proc. of 5th International Conference on Soft Computing and Information/Intelligent Systems (IIZUKA”98)     284 - 287  1998年10月

  • ロボットの前言語的表現とコミュニケーション

    第16回日本ロボット学会学術講演会   PP645-646  1998年09月

  • ロボットにおける感覚行動マップと行動計画への展開

    第16回日本ロボット学会学術講演会   PP429-430  1998年09月

  • ロボットにおける感覚行動マップと行動計画への展開

    尾形哲也, 日塔潔, 菅野重樹

    第16回日本ロボット学会学術講演会   PP429-430   429 - 430  1998年09月

  • ロボットの前言語的表現とコミュニケーション

    尾形哲也, 松山佳彦, 大塚卓美, 菅野重樹

    第16回日本ロボット学会学術講演会   PP645-646   645 - 646  1998年09月

  • ロボットの身体性に基づく感情モデルと内部表象獲得モデル

    日本機械学会ロボティクスメカトロニクス講演会   2CII4-3  1998年06月

  • ロボットの自律系モデルによる情緒表現と評価実験

    日本機械学会ロボティクスメカトロニクス講演会   2BII1-6  1998年06月

  • ロボットの身体性に基づく感情モデルと内部表象獲得モデル

    尾形哲也, 大塚卓美, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会   2CII4-3   2CII4 - 3  1998年06月

  • ロボットと人間の情緒交流—自律系モデルの効果と試行実験

    第3回ロボティクス・シンポジア   PP41-46  1998年05月

  • ロボットと人間の情緒交流—自律系モデルの効果と試行実験

    尾形哲也, 菅野重樹

    第3回ロボティクス・シンポジア   PP41-46   41 - 46  1998年05月

  • Acquisition of Holophrastic Speech in Autonomous Robots-Toward the Emergence of Verbal Communication in Robots-

    Toward a Science of Consciousness "Tucson III   250  1998年04月

  • Acquisition of Holophrastic Speech in Autonomous Robots-Toward the Emergence of Verbal Communication in Robots

    Tetsuya OGATA, Shigeki SUGANO

    Toward a Science of Consciousness "Tucson III"     250  1998年04月

  • 人間共存ロボットにおける心的コミュニケーション

    菅野重樹, 渋谷恒司, 尾形哲也

    電気学会全国大会講演論文集     S18 - 21  1998年03月

  • ロボットの自律系モデルによる情緒表現と評価実験

    尾形哲也, 大塚卓美, 菅野重樹

    日本機械学会ロボティクスメカトロニクス講演会   2BII1-6   2BII1 - 6  1998年03月

  • 行動型自律ロボットWAMOEBA-2と人間との情緒的インタラクション実験

    菅野重樹, 渋谷恒司, 尾形哲也

    第3回重点領域研究「知能ロボット」シンポジウム予稿集     149 - 152  1998年01月

  • Communication between behavior-based robots with emotion model and humans

    T. Ogata, Sh Sugano

    Proceedings of the IEEE International Conference on Systems, Man and Cybernetics   2   1095 - 1100  1998年

     概要を見る

    This study discusses the communication between autonomous robots and humans through the development of a robot which has an emotion model. The model refers to the internal secretion system of humans and it has four kinds of the hormone parameters to use to adjust various internal conditions such as motor output, cooling fan output and sensor gain. We surveyed 126 visitors at '97 International Robot Exhibition held in Tokyo, Japan (Oct. 1997) in order to evaluate psychological impressions of the robot. As a result, the human friendliness of the robot was confirmed and some factors of the human-robot emotional communication were discovered.

  • Communication between behavior-based robots with emotion model and humans.

    Tetsuya Ogata, Shigeki Sugano

    1998 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS, VOLS 1-5     1095 - 1100  1998年  [査読有り]

     概要を見る

    This study discusses the communication between autonomous robots and humans through the development of a robot which has an emotion model. The model refers to the internal secretion system of humans and it has four kinds of the hormone parameters to use to adjust various internal conditions such as motor output, cooling fan output and sensor gain. We surveyed 126 visitors at '97 International Robot Exhibition held in Tokyo, Japan (Oct. 1997) in order to evaluate psychological impressions of the robot. As a result, the human friendliness of the robot was confirmed and some factors of the human-robot emotional communication were discovered.

    DOI

  • シンセシスのための心的言語処理

    尾形哲也, 菅野重樹

    日本学術振興会未来開拓事業人間指向シンセシスの科学 早稲田慶応成果報告会(第1回) 講演論文集     9 - 10  1997年12月

  • ロボットにおける自律的情動反応の生成

    尾形哲也, 菅野重樹

    第15回ロボット学会学術講演会予稿集   2   385 - 386  1997年09月

  • 自律ロボットの人間との対話表現の獲得

    尾形哲也, 大塚卓美, 菅野重樹

    第15回ロボット学会学術講演会予稿集   2   385 - 386  1997年09月

  • 異種ロボット間のインタラクション試行実験

    尾形哲也, 川端邦明, 藤井輝夫, 淺間 一, 遠藤 勲, 菅野重樹

    第15回ロボット学会学術講演会予稿集   3   833 - 834  1997年09月

  • Emergence of emotional expression of robots based on the internal secretion system

    Tetsuya OGATA, Shigeki SUGANO

    The Brain and Self Workshop: Toward a Science of Consciousness   61  1997年08月

  • 人間と機械の心的交流

    菅野重樹, 尾形哲也

    計測自動制御学会学術講演会予稿集     1327 - 1328  1997年08月

  • 自己保存に基づくロボットの行動生成-方法論と機械モデルの実体化

    尾形哲也, 菅野重樹

    日本ロボット学会誌   15 ( 5 ) 710 - 721  1997年07月

  • ロボットにおける自律系?WAMOEBA-2の自律系ハードウェアとその影響?

    尾形哲也, 菅野重樹

    日本機械学会 ロボティクス・メカトロニクス講演会”97 予稿集A     567 - 568  1997年06月

  • ニューラルネットワークによるビヘイビアネットワークの自己組織化

    尾形哲也, 北岸郁雄, 菅野重樹

    日本機械学会 ロボティクス・メカトロニクス講演会”97予稿集B     1169 - 1172  1997年06月

  • ロボットにおける自律系の導入?ハードウェア構成とアルゴリズムの提案?

    菅野重樹, 尾形哲也, 渋谷恒司

    第2回重点領域研究「知能ロボット」シンポジウム予稿集     111 - 114  1997年01月

  • Mechanical systems for autonomic nervous system in robots

    Tetsuya Ogata, Shigeki Sugano

    IEEE/ASME International Conference on Advanced Intelligent Mechatronics, AIM   113   113  1997年

     概要を見る

    The similarities between the human autonomic nervous system and the hardware mechanisms of robots are discussed. The hardware of the autonomic nervous system incorporated into the independent autonomous mobile robot WAMOEBA-2 (Waseda Artificial Mind On Emotion BAse) which has been developed for the realization of the emotional communication between humans and robots is presented. WAMOEBA-2 has four kinds of hormone parameters using the original algorithm, the evaluation function for self-preservation. These parameters influence various parts of the mechanical system so that WAMOEBA-2 can cope with its environment and `express emotions'.

  • Generation of behavior automaton on neural network

    Tetsuya Ogata, Kazuki Hayashi, Ikuo Kitagishi, Shigeki Sugano

    IEEE International Conference on Intelligent Robots and Systems   2   608 - 613  1997年  [査読有り]

     概要を見る

    To plan behavior procedures, it is necessary for an agent to have a world model concerning the temporal sequences information. In this paper, a temporal information learning algorithm is proposed with a three layer neural network implemented the `effectiveness of simulation accumulation' algorithm. This algorithm can construct `behavior automaton' in the neural network. From the results of some learning experiments using a mobile robot simulation, the generated automaton express the complexity of the simulation environments. The robot agent acquires a behavior automaton for obstacle avoidance behavior which is influenced by the simulation environment.

    DOI

  • Emotional behavior adjustment system in robots

    Tetsuya Ogata, Shigeki Sugano

    Robot and Human Communication - Proceedings of the IEEE International Workshop     352 - 357  1997年  [査読有り]

     概要を見る

    To cope with dynamically changing environments, creatures have not only `motor nervous system' but `autonomic nervous system' as an internal adjustment mechanism. However conventional robots have been equipped with only the motor nervous system. We developed an autonomous mobile robot implemented a internal adjustment model based on self-preservation. This model involves the original system named `Behavior-Adjustment System', which influences a whole condition of the robot. This paper describes the concrete methodology for application of the behavior-adjustment system to robots. We confirmed the feasibility of the proposed system for the mobile robot in two practical experiments; `obstacle avoidance problem' and `effect of the voltage of battery'.

  • ロボットにおける自律系の機構

    尾形哲也, 中村好伸, 菅野重樹

    第14回ロボット学会学術講演会予稿集     389 - 390  1996年11月

  • ロボットの自律的行動計画の生成?神経回路による時系列オートマトンの獲得とその応用?

    尾形哲也, 林一樹, 大塚卓美, 北岸郁雄, 菅野重樹

    第14回ロボット学会学術講演会予稿集   2   699 - 700  1996年11月

  • 人間とロボットの情緒交流に関する研究?情緒の表出とその形態に関する一考察?

    尾形哲也, 中村好伸, 菅野重樹

    第14回ロボット学会学術講演会予稿集   3   1149 - 1150  1996年11月

  • 移動ロボットにおける行動計画のための時系列情報獲得?刺激蓄積効果による時系列ネットワークの生成?

    尾形哲也, 林一樹, 菅野重樹

    日本機械学会ロボティクス・メカトロニクス講演会”96予稿集     133 - 136  1996年06月

  • 人間とロボットの情緒交流に関する研究?評価ロボット"WAMOEBA-2"の設計と製作?

    尾形哲也, 菅野重樹

    日本機械学会ロボティクス・メカトロニクス講演会”96予稿集     449 - 452  1996年06月

  • Emergence of Mind in Robots for Human Interface &#8211; Research Methodology and Robot Model

    Shigeki SUGANO, Tetsuya OGATA

    Proc. of IEEE International Conference on Robotics and Automation (ICRA”96)     1191 - 1196  1996年05月

    DOI

  • 人間共存ロボットにおける情緒の意義

    菅野重樹, 尾形哲也

    日本機械学会,No.96-8,第1回ロボメカ・シンポジア講演論文集     97 - 100  1996年05月

  • Emergence of mind in robots for human interface - research methodology and robot model

    Shigeki Sugano, Tetsuya Ogata

    Proceedings - IEEE International Conference on Robotics and Automation   2   1191 - 1198  1996年  [査読有り]

     概要を見る

    The objective of this work is to develop the technology for human-machine communication through the research of the emergence of mind in mechanical systems. In this paper, the hypothesis about the emergence of mind are proposed. First, the system chart expressing the human brain information processing, and the development of an autonomous mobile robot 'WAMOEBA-IR' (Waseda Artificial Mind On Emotion BAse) are described. The conception of the WAMOEBA-IR design is that robots should have Self-Preservation Evaluation Function. Further more, the method to evaluate the whole system are described from the viewpoint of the animal psychology. As a result of the experiments, WAMOEBA-IR showed specific emotional reactions with color appearances to some situations. WAMOEBA-IR has the sense of values about colors and sounds based on Self-Preservation as the first step of the emergence of mind.

    DOI

  • Sense Syncretic Model toward the Construction of ”Robot-Original-Language

    Tetsuya OGATA, Shigeki SUGANO

    Proc. of IEEE International Workshop on Robot and Human Communication (ROMAN”96)     433 - 438  1996年  [査読有り]

  • 情緒交流ロボットWAMOEBA-2の設計

    菅野重樹, 尾形哲也

    第1回重点領域研究「知能ロボット」シンポジウム予稿集     117 - 120  1996年01月

  • ロボットが心を持つ可能性に関する研究?内外感覚の統合?

    尾形哲也, 菅野重樹

    第13回ロボット学会学術講演会予稿集     1195 - 1196  1995年11月

  • ロボットにおける自律系と移動方向判断アルゴリズム

    尾形哲也, 菅野重樹

    第13回ロボット学会学術講演会予稿     265 - 266  1995年11月

  • 人間とロボットの情緒交流に関する研究?ロボットの内部状態の表出?

    尾形哲也, 菅野重樹

    日本機械学会ロボティクス・メカトロニクス講演会”95予稿集     877 - 878  1995年06月

  • ロボットにおける心の発生?志向性と行動?

    尾形哲也, 菅野重樹

    日本機械学会第72期通常総会講演論文集(IV)     271 - 272  1995年04月

  • ロボットにおける心の発生?記憶情報処理の導入?

    尾形哲也, 山本健次郎, 菅野重樹

    第12回ロボット学会学術講演会予稿集     129 - 130  1994年11月

  • ロボットにおける心の発生?第1報:評価用機械モデルの設計・製作?

    菅野重樹, 玉本淳一, 山本健次郎, 尾形哲也, 加藤一郎

    第11回ロボット学会学術講演会予稿集     763 - 766  1993年11月

  • ロボットにおける心の発生?第2報:自己保存評価関数と基礎実験?

    菅野重樹, 玉本淳一, 山本健次郎, 尾形哲也, 加藤一郎

    第11回ロボット学会学術講演会予稿集     767 - 770  1993年11月

▼全件表示

書籍等出版物

  • 「こころ」とアーティフィシャル・マインド

    河合, 俊雄, 吉岡, 洋, 西垣, 通, 尾形, 哲也, 長尾, 真

    創元社  2021年02月 ISBN: 9784422117577

  • 発達ロボティクスハンドブック : ロボットで探る認知発達の仕組み

    Cangelosi, Angelo, Schlesinger, Matthew, 萩原, 良信, 荒川, 直哉, 長井, 隆行, 尾形, 哲也, 稲邑, 哲也, 岩橋, 直人, 杉浦, 孔明, 牧野, 武文, 岡田, 浩之, 谷口, 忠大

    福村出版  2019年01月 ISBN: 9784571230592

  • ディープラーニングがロボットを変える

    尾形, 哲也

    日刊工業新聞社  2017年07月 ISBN: 9784526077326

Misc

  • Deep Neural Networkを用いたマルチモーダル音声認識

    野田邦昭, 山口雄紀, 中臺一博, 奥乃博, 尾形哲也

    日本ロボット学会学術講演会予稿集(CD-ROM)   32nd   ROMBUNNO.1I1-04  2014年09月

    J-GLOBAL

  • 人間の描画発達に基づくロボットの描画模倣学習モデルの構築

    西出俊, 望月敬太, 奥乃博, 尾形哲也

    日本ロボット学会学術講演会予稿集(CD-ROM)   32nd   ROMBUNNO.2I2-04  2014年09月

    J-GLOBAL

  • Robust Multipitch Analyzer against Initialization based on Latent Harmonic Allocation using Overtone Corpus

    Daichi Sakaue, Katsutoshi Itoyama, Tetsuya Ogata, HiroshiG.Okuno

    情報処理学会論文誌   54 ( 4 )  2013年04月

     概要を見る

    We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed "overtone corpus." Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.21(2013) No.2 (online)DOI http://dx.doi.org/10.2197/ipsjjip.21.246------------------------------We present a Bayesian analysis method that estimates the harmonic structure of musical instruments in music signals on the basis of psychoacoustic evidence. Since the main objective of multipitch analysis is joint estimation of the fundamental frequencies and their harmonic structures, the performance of harmonic structure estimation significantly affects fundamental frequency estimation accuracy. Many methods have been proposed for estimating the harmonic structure accurately, but no method has been proposed that satisfies all these requirements: robust against initialization, optimization-free, and psychoacoustically appropriate and thus easy to develop further. Our method satisfies these requirements by explicitly incorporating Terhardt's virtual pitch theory within a Bayesian framework. It does this by automatically learning the valid weight range of the harmonic components using a MIDI synthesizer. The bounds are termed "overtone corpus." Modeling demonstrated that the proposed overtone corpus method can stably estimate the harmonic structure of 40 musical pieces for a wide variety of initial settings.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.21(2013) No.2 (online)DOI http://dx.doi.org/10.2197/ipsjjip.21.246------------------------------

    CiNii

  • マイク数以上の同時発話分離のための調波・非調波音源モデルの検討

    平澤恭治, 安良岡直希, 高橋徹, 尾形哲也, 奥乃博

    第74回全国大会講演論文集   2012 ( 1 ) 577 - 578  2012年03月

     概要を見る

    人間の生活環境に存在する多数の音源を正しく分離するために, 我々はマイク数以上の音源分離(劣決定音源分離)の検討を行っている. 我々は同時発話分離のために混合ガウス分布を用いた調波・非調波音源モデルを提案しているが, 調波部分は理論的に正当化されたモデル化が行われている一方で, 非調波部分のモデルには理論的な正当性がなく, 分離性能を低下させる要因となっていた. 本稿では従来の調波・非調波音源モデルに対して変更を加え, SiSEC2011で用いられた男女3-4話者の混合音声を分離する実験により, どのようなモデルが劣決定同時発話分離問題に適しているかを検討する.

    CiNii

  • マイク数以上の同時発話分離のための調波・非調波音源モデルの検討

    平澤恭治, 安良岡直希, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会全国大会講演論文集   74th ( 2 ) 2.577-2.578 - 579  2012年03月

     概要を見る

    人間の生活環境に存在する多数の音源を正しく分離するために, 我々はマイク数以上の音源分離(劣決定音源分離)の検討を行っている. 我々は同時発話分離のために混合ガウス分布を用いた調波・非調波音源モデルを提案しているが, 調波部分は理論的に正当化されたモデル化が行われている一方で, 非調波部分のモデルには理論的な正当性がなく, 分離性能を低下させる要因となっていた. 本稿では従来の調波・非調波音源モデルに対して変更を加え, SiSEC2011で用いられた男女3-4話者の混合音声を分離する実験により, どのようなモデルが劣決定同時発話分離問題に適しているかを検討する.

    CiNii J-GLOBAL

  • 同時複数音源に対する擬音語による音源選択システム

    山村祐介, 高橋徹, 尾形哲也, 奥乃博

    情報処理学会全国大会講演論文集   74th ( 2 ) 2.587-2.588  2012年03月

    J-GLOBAL

  • MAHL:演奏者間のインタラクション分析のためのスコアアライメント手法の提案

    前澤 陽, 糸山 克寿, 尾形 哲也, 奥乃 博

    研究報告音楽情報科学(MUS)   2011 ( 19 ) 1 - 6  2011年07月

     概要を見る

    本稿では、楽器パート毎に、楽譜と音響信号のアライメントを算出する手法を提案する。本手法では、各楽器パートに共通の、自己回帰過程に従うテンポモデルを持たせる。各楽器パートの時系列は隠れセミマルコフモデルに従い、状態継続長の事前分布としてテンポモデルを持つ。また、音響信号の出力は潜在的調波配分法に従う。パート間の揺らぎを持たせない場合の、アライメントの性能を評価し、アライメント手法としての有用性が確認された。また、演奏における発音タイミングの揺らぎがモデル化できることが示唆された。This paper presents a method to align an audio signal and individual music instrument parts comprising a music score. Such method allows a machine to analyze temporal interaction of music performers. Proposed method is based on fitting multiple Hidden Semi-Markov Models (HSMM) to the observed audio signal, each HSMM of which emits Latent Harmonic Allocation parameters. Each HSMM corresponds to a music instrument part, and the state duration probability is conditioned on an auto-regressive tempo model. Evaluation suggests usefulness as score alignment method, and hints at the usefulness as multiple part alignment method.

    CiNii

  • 2A1-D09 自己形態主張カスタマイズロボットの開発 : ユーザの自己効力感向上がインタラクションに与える効果の検証(コミュニケーション-ロボット)

    山崎 由美子, 守 良真, 菅 祐樹, 尾形 哲也, 菅野 重樹

    ロボティクス・メカトロニクス講演会講演概要集   2011   "2A1 - D09(1)"-"2A1-D09(2)"  2011年05月

     概要を見る

    Our goal is to crate a robot which can communicate with people for a long time. To make this possible, we are developing a customizable communication robot "WEAR (Waseda Extendable ARchitecture)". We also provided modules, and experimental subjects can customize the robot by inserting them into the module-port. At the same time, the robot can recognize the customize, and it also can express its autonomy by rejecting the customization by ejecting the module. In this paper, we proposed the communication model to grow self-efficacy. And as the result of the experiments, we found that the proposed communication model is preferred by user than all-accept communication model.

    CiNii

  • 発語行為レベルの情報を用いた音声対話システムの構築とデータ分析

    松山匡子, 駒谷和範, 武田龍, 尾形哲也, 奥乃博

    人工知能学会言語・音声理解と対話処理研究会資料   61st   7 - 12  2011年03月

    J-GLOBAL

  • 拍長の連続性を考慮した潜在的調波配分法に基づくスコアアライメント手法

    前澤陽, 後藤真孝, 尾形哲也, 奥乃博

    日本音響学会研究発表会講演論文集(CD-ROM)   2011   ROMBUNNO.3-1-15  2011年03月

    J-GLOBAL

  • 予測可能性による身体識別及び身体図式獲得

    信田春満, 日下航, 尾形哲也, 高橋徹, 奥乃博

    情報処理学会全国大会講演論文集   73rd ( 2 ) 2.125-2.126  2011年03月

    J-GLOBAL

  • バージイン許容音声対話システムにおけるユーザ発話の分析と指示対象同定への応用

    松山 匡子, 駒谷 和範, 武田 龍, 尾形 哲也, 奥乃 博

    研究報告音声言語情報処理(SLP)   2010 ( 21 ) 1 - 6  2010年07月

     概要を見る

    本稿は,バージイン許容列挙型音声対話におけるユーザ発話の分析と,分析結果を応用した指示対象同定手法の拡張について報告する.バージイン許容音声対話では,個々のユーザやシステムの発話内容によってユーザの発話タイミングや発話表現が異なる.そこでこれらを事前確率として反映させ,発話意図解釈の性能向上を図る.我々はまず,ニュース読み上げとクイズの2つの列挙型対話システムで収集したユーザ発話 1584 発話を分析し,ユーザの参照表現発話率が個々のユーザやシステムの列挙項目長に依存することを明らかにした.さらに,これらの特性を指示対象同定の枠組みに組み込み,タイミングと音声認識結果の解釈の事前確率として反映させる.この事前確率の推定には,ロジスティック回帰を用いる.事前確率として一定値を用いた場合に比べて,指示対象同定精度が最大 6.2 ポイント向上することを実験により確認した.This paper reports the extension of identification method based on analyses of user utterance in barge-in-able spoken dialogue system which reads out items. Generally, user&#039;s behaviors such as barge-in timing and utterance expressions vary in accordance with the user&#039;s preference and the content of system utterances. To interpret users&#039; intention robustly, first, we analyze 1584 utterances collected by our systems with quiz and news-listing tasks and reveal that the ratio of using referential expressions depends on individual users and average lengths of listed items. Second, we incorporate this tendency as a prior probability into our probabilistic framework for identifying user&#039;s intended item. This prior probability is calculated by logistic regression. Experimental results show that our method improves the identification accuracy by as many as 6.2 points in the best case over the non-informative prior.

    CiNii

  • 複数の言語モデルと言語理解モデルによるラピッドプロトタイピング向け音声理解

    勝丸真樹, 駒谷和範, 中野幹生, 船越孝太郎, 辻野広司, 尾形哲也, 奥乃博

    情報処理学会研究報告(CD-ROM)   2009 ( 6 ) ROMBUNNO.SLP-80,5  2010年04月

    J-GLOBAL

  • クラシック音楽理解力拡張インタフェースを目指して : 複数の演奏家による解釈共通旋律と解釈相違旋律の推定

    前澤 陽, 後藤 真孝, 高橋 徹, 駒谷 和範, 尾形 哲也, 奥乃 博

    全国大会講演論文集   72 ( 0 ) 143 - 144  2010年03月

    CiNii J-GLOBAL

  • 複数の言語モデルと言語理解モデルによるラピッドプロトタイピング向け音声理解

    勝丸 真樹, 駒谷 和範, 中野 幹生, 船越 孝太郎, 辻野 広司, 尾形 哲也, 奥乃 博

    研究報告音声言語情報処理(SLP)   2010 ( 5 ) 1 - 6  2010年02月

     概要を見る

    本稿では,少量の学習データでも高精度な音声理解を実現する手法について述べる.学習データが少ない場合,単一の音声理解方式による精度は低い傾向にある.そこで本手法では,まず,複数の言語モデルと言語理解モデルを用いて複数の理解結果を得ることで,対処可能な発話を増やす.次に,得られた複数の理解結果に対して,ロジスティック回帰に基づき発話単位の信頼度を付与し,その信頼度が最も高い理解結果を選択する.ロジスティック回帰には,学習データ増加時の回帰係数の変化量に着目することで,必要最低限の学習データを割り当てる.評価実験では,学習データが少ない場合でも,単一の音声理解方式と比較して,本手法が高い音声理解精度を得られることを示す.We aim to improve a speech understanding module with a small amount of training data. High performance is not obtained by single speech understanding methods especially when the amount of available training data is small. We utilize multiple language models (LMs) and language understanding models (LUMs) to cover various user utterances. Then, the most appropriate speech understanding result is selected from several candidates on the basis of confidence measures calculated by logistic regressions. We determine necessary amount of training data for the regressions by focusing on changes in their coefficients when the training data increases. We evaluate our method for various amounts of training data and confirm that our method outperforms every single speech understanding method even when only a small amount of training data is available.

    CiNii

  • 音響信号とコンテキスト制約を併用したバイオリンの演奏弦系列の推定

    前澤陽, 糸山克寿, 高橋徹, 尾形哲也, 奥乃博

    日本音響学会研究発表会講演論文集(CD-ROM)   2009   ROMBUNNO.2-5-15  2009年09月

    J-GLOBAL

  • バージイン発話タイミングモデルを導入した指示対象同定

    松山匡子, 駒谷和範, 武田龍, 尾形哲也, 奥乃博

    情報処理学会研究報告(CD-ROM)   2009 ( 1 ) ROMBUNNO.NL-191,14  2009年06月

    J-GLOBAL

  • バージイン発話タイミングモデルを導入した指示対象同定

    松山 匡子, 駒谷 和範, 武田 龍, 尾形 哲也, 奥乃 博

    研究報告自然言語処理(NL)   2009 ( 14 ) 1 - 7  2009年05月

     概要を見る

    自然な会話を実現できる音声対話システムでは,ユーザが自由なタイミングや言語表現で発話できることが望ましい.我々は,ユーザが任意のタイミングでシステム発話に割り込み(バージイン)できる手法を開発している.本手法では,Independent Component Analysis (ICA) に基づくセミブラインド音源分離を利用している.本稿では,システムが列挙する項目に対してユーザがバージイン発話で指定した対象を同定するために,ユーザのバージイン発話から得られるタイミング情報を用いて解釈する新手法について報告する.まず,ユーザが参照表現を用いて発話する場合のタイミング分布を,予備調査の結果に基づき,ガンマ分布で近似する.次に,システムの読み上げる各項目に対して,ユーザ発話がそのタイミングで解釈されるべき場合とその音声認識結果で解釈されるべき場合とをそれぞれ確率として表現する.これら2つの確率を統合し,最も尤度の高い項目をユーザの指示対象と同定する.システムが列挙する項目の一つを指定するユーザのバージイン発話400発話に対して,本手法が2つのベースライン手法(音声認識結果のみから指示対象を同定する手法,及び,ユーザの発話タイミングのみから指示対象を同定する手法) よりも高精度に同定できることを実験により確認した.In conversational dialogue systems, the user prefers to speak at any time and to use natural expressions. We have developed an Independent Component Analysis (ICA) based semiblind source separation method, which allows users to barge-in over system utterances at any time. We create a novel method from timing information derived from barge-in utterances to identify one item that a user indicates during system enumeration. First, we determine the timing distribution of user utterances containing referential expressions and then approximate it using gamma distribution. Second, we represent both the utterance timing and automatic speech recognition (ASR) results as probabilities of the desired selection from the system&#039;s enumeration. We then integrate these two probabilities to identify the item having the maximum likelihood of selection. Experimental results using 400 utterances indicated that our method outperformed two methods used as a baseline (one of ASR results only and one of utterance timing only) in identification accuracy.

    CiNii

  • 神経回路モデルを用いた音声模倣モデルによる音声バブリングと母音獲得過程シミュレーション

    神田 尚, 尾形 哲也, 高橋 徹, 駒谷 和範, 奥乃 博

    全国大会講演論文集   71 ( 0 ) 133 - 134  2009年03月

    CiNii J-GLOBAL

  • ベース音高確率とクロマベクトルの相関を考慮した和音進行認識

    高野 秀樹, 須見 康平, 糸山 克寿, 駒谷 和範, 尾形 哲也, 奥乃 博

    全国大会講演論文集   71 ( 0 ) 241 - 242  2009年03月

    CiNii J-GLOBAL

  • 顔の動作に追従したインタフェースを持つ音環境可視化システム

    久保田 祐史, 吉田 雅敏, 駒谷 和範, 尾形 哲也, 奥乃 博

    全国大会講演論文集   70 ( 0 ) 273 - 274  2008年03月

    CiNii J-GLOBAL

  • トピック推定と対話履歴の統合によるドメイン選択を行うマルチドメイン音声対話システム

    池田 智志, 駒谷 和範, 尾形 哲也, 奥乃 博

    全国大会講演論文集   70 ( 0 ) 139 - 140  2008年03月

    CiNii J-GLOBAL

  • 音声対話システムにおける動的ヘルプ生成を指向したWFSTに基づく文法検証によるユーザ知識推定

    福林雄一朗, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会言語・音声理解と対話処理研究会資料   52nd   45 - 50  2008年02月

    J-GLOBAL

  • 2P1-G03 Segmenting Sound Signals and Articulatory Movement using Recurrent Neural Network toward Phoneme Acquisition

    神田 尚, 尾形 哲也, 駒谷 和範, 奥乃 博

    ロボティクス・メカトロニクス講演会講演概要集   2008 ( 0 ) _2P1 - G03_1-_2P1-G03_4  2008年

     概要を見る

    This paper proposes a computational model for phoneme acquisition by infants. Infants perceive speech not as discrete phoneme sequences but as continuous acoustic signals. One of critical problems in phoneme acquisition is the design for segmenting these continuous speech. The key idea to solve this problem is that articulatory mechanisms such as the vocal tract help human beings to perceive sound units corresponding to phonemes. To segment acoustic signal with articulatory movement, our system was implemented by using a physical vocal tract model, called the Maeda model, and applying a segmenting method using Recurrent Neural Network with Parametric Bias (RNNPB). This method determines segmentation boundaries in a sequence using the prediction error of the RNNPB model, and the PB values obtained by the method can be encoded as kind of phonemes. Experimental results demonstrated that our system could self-organize the same phonemes in different continuous sounds. This suggests that our model reflects the process of phoneme acquisition.

    CiNii

  • バージインを許容するロボット音声対話のためのICAを用いたセミブラインド音源分離

    武田龍, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報科学技術フォーラム   FIT 2007   261 - 262  2007年08月

    J-GLOBAL

  • LE-008 音声認識結果とコンセプトへの重みづけによるWFSTに基づく音声言語理解の高精度化(自然言語・音声・音楽)

    福林 雄一朗, 駒谷 和範, 中野 幹生, 船越 孝太郎, 尾形 哲也, 奥乃 博

    情報科学技術レターズ   6   133 - 134  2007年08月

    CiNii

  • 楽曲推薦システムの効率性とスケーラビリティの改善のための確率的推薦モデルのインクリメンタル学習法

    吉井和佳, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会研究報告   2007 ( 81(MUS-71) ) 19 - 26  2007年08月

     概要を見る

    本稿では,ハイブリッド型楽曲推薦システムの効率性とスケーラビリティを改善する手法について述べる.我々は以前,確率的推薦モデルを用いて「ユーザの楽曲評価」と「楽曲の音響的特徴」とを同時に考慮するハイブリッド型推薦システムを開発し,高精度な推薦ができることを実証したが,効率性とスケーラビリティに関する問題が残されていた.すなわち,新規評価や新規ユーザ,新規楽曲が追加されるたびにモデル全体を再学習する必要があった.また,モデル学習にかかる計算コストがユーザ数と楽曲数に比例して増加するため,現実的なサイズのデータを扱うことができなかった.そこで本稿では,効率性を改善するため,モデルをわずかなコストで部分的に更新できるインクリメンタル学習法を提案する.さらに,スケーラビリティを改善するため,まず少数の代表的な仮想ユーザと仮想楽曲に対して「コアモデル」を構築し,次にインクリメンタル学習法によって実際のユーザと楽曲をコアモデルに追加登録していく手法を提案する.実験の結果,上記二つの改良を施した推薦システムでは,推薦精度が向上することが分かった.We aimed at improving the efficiency and scalability of a hybrid music recommender system. Although this system was proved to make accurate recommendations by using a probabilistic model that integrates rating scores provided by users acoustic features of musical pieces, it lacks efficiency and scalability. That is, the entire model needs to be re-trained from scratch whenever a new score, user, or piece is added. Furthermore, the system cannot deal with practical numbers of users and pieces. To improve efficiency, we propose an incremental method that partially updates the model at low computational cost. To enhance scalability, we propose a method that first constructs a small &quot;core&quot; model over fewer virtual representatives created from real users and pieces, and then adds the real users and pieces to the core model by using the incremental method. The experimental results revealed that the proposed system was not only efficient and scalable but also outperformed the original system in terms of accuracy.

    CiNii J-GLOBAL

  • 可聴音波を用いたAHによる遮蔽物の検出と距離推定法

    丹羽治彦, 尾形哲也, 駒谷和範, 奥乃博

    日本音響学会研究発表会講演論文集(CD-ROM)   2007   1-10-7  2007年03月

    J-GLOBAL

  • 音を視覚化する録音再生システム

    吉田雅敏, 海尻聡, 山本俊一, 中臺一博, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会全国大会講演論文集   69th ( 2 ) 2.577-2.578  2007年03月

    J-GLOBAL

  • 自己身体モデルの投影に基づく模倣行為中における他者の発見

    横矢龍之介, 尾形哲也, 谷淳, 駒谷和範, 奥乃博

    情報処理学会全国大会講演論文集   69th ( 2 ) 2.445-2.446  2007年03月

    J-GLOBAL

  • 可聴音波を用いたAHによる遮蔽物の検出と距離推定法

    丹羽治彦, 尾形哲也, 駒谷和範, 奥乃博

    電子情報通信学会技術研究報告   106 ( 267(EA2006 48-54) ) 1 - 6  2006年09月

     概要を見る

    本研究では、可聴域の音波の回折現象を利用してオクルージョン問題における遮蔽された物体の検出とその距離を計測する手法について報告する。まず可聴域音波を使用した場合の距離計測の性能について調査した。計測手法は基本的に超音波距離計測で用いられるTOF法に基づいた相互相関法を用いた。次に2物体を同軸かつ正面に配置した場合の同手法による同時距離推定を検証した。その問題点として背後の物体からの反射波信号のSNRが小さいというものがあった。そこで背後の物体からの反射波を強調して獲得するために、3次元音場解析法であるAH法を適用した。その結果背後物体からの反射信号のSNRの増加を可能にした。本手法は奥行き1.2mの範囲において同サイズの2物体に対して有効であり、その距離推定精度は正面の物体には3cm程度、背後の物体には6cm程度であった。

    CiNii J-GLOBAL

  • Instrogramを用いた楽器構成に基づく類似楽曲検索

    北原鉄朗, 後藤真孝, 駒谷和範, 尾形哲也, 奥乃博

    日本音響学会研究発表会講演論文集(CD-ROM)   2006   2-7-13  2006年09月

    J-GLOBAL

  • lnstrogram:発音時刻検出とFO推定の不要な楽器音認識手法

    北原 鉄朗, 後藤 真孝, 駒谷 和範, 尾形 哲也, 奥乃 博

    情報処理学会研究報告音楽情報科学(MUS)   2006 ( 90 ) 69 - 76  2006年08月

     概要を見る

    本稿では,Instrogramと呼ばれる楽器存在確率の時間・周波数表現に基づく新たな楽器音認識手法を提案する.従来の多くの楽器音認識では単音を処理単位としていたため,各単音の発音時刻や基本周波数(FO)を正確に推定する必要があった.しかし,混合音におけるそれらの推定は難しく,推定誤りによって楽器音認識精度は大きく低下していた.本手法では,これらの推定をせずに,あらゆるFOに関して楽器存在確率の時系列を求め,Instrogramと呼ばれるスペクトログラムのような視覚表現として可視化する.実験の結果,Instrogramが実際の楽器構成を表していることを確認した.さらに,Instrogram間の類似度計算に基づいた類似楽曲検索も実現した.This paper describes a new musical instrument recognition method based on the time-frequency representation of instrumentation called an instrogram. Because the conventional instrument recognition is performed for each note, accurate estimation of the onset time and fundamental frequency (FO) is required. These estimation is, however, difficult in polyphonic music, and thus their errors deteriorated the recognition performance severely. Without these estimation , our method calculates the temporal trajectory of instrument existence probabilities for every possible FO and visualizes them as a spectrogram-like graphical representation,called an instrogram. Experimental results show that instrograms represent actual instrumentation. We have also achieved music information retrieval by calculating the similarity between instrograms.

    CiNii

  • 音楽音響信号と歌詞の時間的対応付け手法:歌声の分離と母音のViterbiアラインメント

    藤原弘将, 後藤真孝, 緒方淳, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会研究報告   2006 ( 90(MUS-66) ) 37 - 44  2006年08月

     概要を見る

    本稿では,伴奏音を含む音楽音響信号と対応する歌詞の時間的な対応付け手法について述べる.クリーンな音声信号とその発話内容の時間的対応付けを推定をするViterbi アラインメント手法はこれまでも存在したが,歌声と同時に演奏される伴奏音の悪影響で市販 CD 中の歌声には適用できなかった.本稿では,この問題を解決するため,歌声の調波構造を抽出・再合成することで混合音中の歌声を分離する手法,歌声・非歌声状態を行き来する隠れマルコフモデル (HMM)を用いた歌声区間検出手法,音響モデルを分離歌声に適応させることで Viterbi アラインメントを適用する手法を提案する.日本語のポピュラー音楽を用いた評価実験を行い,本手法により10曲中8曲について十分な精度で音楽と歌詞の対応付けが出来ることを確かめた.This paper describes a method that can automatically synchronize between polyphonic musical audio signals and corresponding lyrics. Although there were methods that can synchronize between monophonic speech signals and corresponding text transcriptions by using Viterbi alignment techniques, they cannot be applied to vocals in CD recordings because accompaniment sounds often overlap with vocals. To align lyrics with such vocals, we therefore developed three methods: a method for segregating vocals from polyphonic sound mixtures by extracting and resynthesizing the vocal melody, a method for detecting vocal sections using a Hidden Markov Model (HMM) that transitions back and forth between vocal and non-vocal state, and a method for adapting a speech-recognizer phone model to segregated vocal signals. Experimental results for 10 Japanese popular-music songs showed that our system can synchronize between music and lyrics with satisfactory accuracy for 8 songs.

    CiNii J-GLOBAL

  • ICAによる音源分離とミッシングフィーチャマスクの自動生成による同時発話認識

    武田龍, 山本俊一, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会全国大会講演論文集   68th ( 2 ) 343 - 344  2006年03月

    J-GLOBAL

  • Dynamic Communication of Humanoid Robot with Multiple People Based on Interaction Distance

    Tasaki Tsuyoshi, Matsumoto Shohei, Ohba Hayato, Yamamoto Shunichi, Toda Mitsuhiko, Komatani Kazunori, Ogata Tetsuya, Okuno Hiroshi G

    Information and Media Technologies   1 ( 1 ) 285 - 295  2006年

     概要を見る

    Research on human-robot interaction is getting an increasing amount of attention. Since most research has dealt with communication between one robot and one person, quite few researchers have studied communication between a robot and multiple people. This paper presents a method that enables robots to communicate with multiple people using the &amp;ldquo;selection priority of the interactive partner&amp;rdquo; based on the concept of &lt;i&gt;Proxemics&lt;/i&gt;. In this method, a robot changes active sensory-motor modalities based on the &lt;i&gt;interaction distance&lt;/i&gt; between itself and a person. Our method was implemented into a humanoid robot, &lt;i&gt;SIG2&lt;/i&gt;. &lt;i&gt;SIG2&lt;/i&gt; has various sensory-motor modalities to interact with humans. A demonstration of &lt;i&gt;SIG2&lt;/i&gt; showed that our method selected an appropriate interaction partner during interaction with multiple people.

    CiNii

  • ヒューマノイドを対象としたミッシングフィーチャー理論による分離音の連続音声認識の向上

    山本俊一, VALIN Jean‐Marc, 中台一博, 中野幹生, 辻野広司, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会学術講演会予稿集(CD-ROM)   23rd   3C35  2005年09月

    J-GLOBAL

  • ロボツトとの協調作業のためのRNNによる擬似シンボルの獲得

    大谷拓, 大庭隼人, 駒谷和範, 尾形哲也, 谷淳, 奥乃博

    情報処理学会全国大会講演論文集   67th ( 2 ) 229 - 230  2005年03月

    J-GLOBAL

  • ミッシングフィーチャ理論を適用した同時発話認識システムの同時発話文による評価

    山本俊一, VALIN Jean‐Marc, 中台一博, 中野幹生, 辻野広司, 駒谷和範, 尾形哲也, 奥乃博

    人工知能学会AIチャレンジ研究会   22nd   101 - 106  2005年

    J-GLOBAL

  • ヒューマノイドSIG2の近接学に基づく複数の人とのインタラクション

    田崎豪, 松本祥平, 大庭隼人, 戸田充彦, 駒谷和範, 尾形哲也, 奥乃博

    日本ロボット学会学術講演会予稿集(CD-ROM)   22nd   1E33  2004年09月

    J-GLOBAL

  • データベース検索音声対話システムにおける履歴を考慮した検索条件の管理

    神田直之, 駒谷和範, 尾形哲也, 奥乃博

    情報科学技術フォーラム   FIT 2004   131 - 132  2004年08月

    J-GLOBAL

  • 和音区間検出と和音名同定の相互依存性を解決する和音認識手法

    吉岡拓也, 北原鉄朗, 駒谷和範, 尾形哲也, 奥乃博

    情報処理学会研究報告   2004 ( 84(MUS-56) ) 33 - 40  2004年08月

    J-GLOBAL

  • 三面図の暖昧性除去における二分決定グラフの利用

    望月敬太, 西出俊, 奥乃博, 尾形哲也

    全国大会講演論文集   49 ( 0 ) 9 - 10  1994年09月

     概要を見る

    三面図は3次元物体の2次元平面上での表現として,歴史的に広く用いられてきた経緯がある.ただ,本来3次元空間に存在するものを2次元平面上へ写像するため,三面図が曖昧になるケースが存在し,その際には対応する3次元モデルが一意に定まらないという不都合が生じる.この不都合を克服するために,対応する3次元モデルを一意に生成するためにヒューリスティクスを用いる研究も存在するが,我々はこれまで三面図の曖昧性の除去に物体表現の観点から取り組んできた.3次元物体のその他の表現法として,位相的に異なるすべての可能な面図を節点とし,それらの間の可能な遷移関係に対して枝を張ったaspect graphと呼ばれるものがある.我々は,aspect graphが有する表現力の完全性を情報量の冗長性をできるだけ避けながら三面図に反映させること,つまり生成用の表現と認識用の表現の統合を目指す観点にたって研究を進めている.なお本稿で扱う3次元物体のクラスは,多面体のうち複数に分離していないもの(多様体)に限るものとする.

    CiNii J-GLOBAL

▼全件表示

産業財産権

  • 情報処理装置、情報処理方法、及びプログラム

    特許第6955733号

    浅谷 学嗣, 桐谷 太郎, 宮崎 祐太, 尾形 哲也

    特許権

    J-GLOBAL

  • 制御装置、システム、学習装置および制御方法

    森 裕紀, 鳥島 亮太, 尾形 哲也, 高橋 城志, 岡野原 大輔

    特許権

    J-GLOBAL

  • 情報処理システムおよび情報処理方法、並びにプログラム

    4472506

    林 隆志, 金 天海, 尾形 哲也

    特許権

    J-GLOBAL

受賞

  • Best Paper Award Finalist

    2022年01月   IEEE/SICE International Symposium on System Integration (SII 2022)   Sensory-Motor Learning for Simultaneous Control of Motion and Force: Generating Rubbing Motion against Uneven Object  

    受賞者: Hiroshi Ito, Takumi Kurata, and Tetsuya Ogata

  • Best paper award

    2022年01月   IEEE/SICE International Symposium on System Integration (SII 2022)   Buttoning Task with a Dual-Arm Robot: An Exploratory Study on a Marker-based Algorithmic Method and Marker-less Machine Learning Methods  

    受賞者: Wakana Fujii, Kanata Suzuki, Tomoki Ando, Ai Tateishi, Hiroki Mori, Tetsuya Ogata

  • Best RoboCup Paper Award Nomination Finalist

    2021年09月   IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2021)   Flexible Object Manipulation for Humanoid Robot Using Partially Binarized Auto-Encoder on FPGA  

    受賞者: Satoshi Ohara, Tetsuya Ogata, and Hiromitsu Awano

  • Best paper award on Cognitive Robotics

    2021年06月   IEEE International Conference on Robotics and Automation (ICRA 2021)   How to select and use tools? : Active Perception of Target Objects Using Multimodal Deep Learning  

    受賞者: Namiko Saito, Tetsuya Ogata, Satoshi Funabashi, Hiroki Mori, Shigeki Sugano

  • SI2020優秀講演賞

    2020年12月   計測自動制御学会SI部門   データ変換に着目したミドルウェアモデルにおけるデバイス類の位置の扱いに関する議論  

    受賞者: 菅佑樹, 森裕紀, 尾形哲也

  • 全国大会優秀賞

    2020年07月   人工知能学会   過去から未来までの文脈を考慮した神経回路モデルによるロボットの目標に基づいた柔軟な行動生成  

    受賞者: 佐藤琢, 村田真悟, 出井勇人, 尾形哲也

  • 全国大会優秀賞

    2020年07月   人工知能学会   未知語に対応可能な言語と動作の統合表現獲得モデル  

    受賞者: 豊田みのり, 森裕紀, 鈴木彼方, 林良彦, 尾形哲也

  • 論文賞

    2019年12月   FA財団  

  • Best paper award

    2019年09月   日本ロボット学会欧文誌   Dynamic Motion Learning for Multi-DOF Flexible-Joint Robots Using Active-Passive Motor Babbling through Deep Learning  

    受賞者: Kuniyuki Takahashi, Tetsuya Ogata, Jun Nakanishi, Gordon Cheng, Shigeki Sugano

  • ROBOMECH表彰(産業・応用分野)

    2019年06月   日本機械学会ロボティクスメカトロニクス部門   深層学習を用いた要素動作の統合手法の開発  

    受賞者: 伊藤洋, 山本健次郎, 尾形哲也

  • IBM 2017 Faculty Award

    2017年07月   IBM  

    受賞者: 尾形哲也

  • Best paper award

    2016年09月  

  • ティーチングアワード総長賞

    2015年03月   早稲田大学  

    受賞者: 尾形哲也, 橋田朋子

  • Best paper award (Robotics)

    2011年12月   IEEE/SICE SI2011  

    受賞者: Kenri Kodaka, Tetsuya Ogata

  • 栢森情報科学振興財団 設立10周年記念・特別研究助成受賞

    2006年02月   栢森情報科学振興財団  

    受賞者: 尾形哲也

  • 日本機械学会論文賞受賞

    2000年03月   日本機械学会  

    受賞者: 尾形哲也, 菅野重樹

▼全件表示

共同研究・競争的資金等の研究課題

  • 一人に一台一生寄り添うスマートロボットAIREC

    内閣府  ムーンショット

    研究期間:

    2020年10月
    -
    2025年03月
     

  • 実世界に埋め込まれる人間中心の人工知能技術の研究開発

    NEDO  人と共に進化する次世代人工知能に関する技術開発事業

    研究期間:

    2020年04月
    -
    2022年03月
     

  • Dense 3-axis tactile sensing and AI to implement human-like manual skills in robots

    日本学術振興会  科学研究費助成事業 基盤研究(B)

    研究期間:

    2019年04月
    -
    2022年03月
     

    シュミッツ アレクサンダー, 尾形 哲也, 玉城 絵美, Somlor Sophon

     概要を見る

    In this research we develop a smart sensing system that enables robot hands to achieve human-like manipulation skills. Key components are 1. dense 3-axis tactile sensors for robot hands and 2. learning algorithms exploiting massive 3-axis tactile data for intelligent force control.
    <BR>
    We integrated the tactile skin sensors in grippers and robot hands. Using a novel joint (with a remote center of motion mechanism) we could achieve full coverage of the palmar side of the fingers with sensors in one gripper. Furthermore, we instrumented human hands with the sensors, to enable skill transfer from human to robot hands in the future. We used the skin sensors integrated in the robot hands for various machine learning experiments. In particular, we used deep convolutional neural networks for tactile object recognition as well as for in-hand manipulation.

  • 記号創発ロボティクスによる人間機械コラボレーション基盤創成

    科学技術振興機構  CREST

    研究期間:

    2015年10月
    -
    2021年03月
     

  • 日常生活支援ロボット

    NEDO  次世代人工知能・ロボット中核技術開発

    研究期間:

    2017年04月
    -
    2020年03月
     

  • 深層学習を用いたロボットの動作プリミティブの獲得と行動生成

    日本学術振興会  科学研究費助成事業 基盤研究(A)

    研究期間:

    2015年04月
    -
    2020年03月
     

    尾形 哲也, 有江 浩明

     概要を見る

    近年,深層学習が多様な領域で利用されているが,その適用範囲は電子化されたデータ処理に特化されており,実世界での作業性が十分に得られていない.一方,ロボットを用いた生活支援が強く期待されている.近年は,汎用型のロボットOSを利用した多機能型汎用ロボットの可能性が着目されている.そこで本研究では,ロボットOSと深層学習を用いたロボット動作学習を実現し,ロボットの知能化と開発コスト削減を行った.
    具体的には,乳幼児の発達学習に関連した認知発達ロボティクス研究の成果を活用し,模倣学習,予測符号化などの概念を元に,実ロボットのから得られる感覚運動情報(経験)を深層学習によりモデル化する手法を提案した.

  • 構成論的発達科学-胎児からの発達原理の解明に基づく発達障害のシステム的理解-

    日本学術振興会  科学研究費助成事業 新学術領域研究(研究領域提案型)

    研究期間:

    2012年06月
    -
    2017年03月
     

    國吉 康夫

  • 社会的認知発達モデルとそれに基づく発達障害者支援システム構成論

    日本学術振興会  科学研究費助成事業 新学術領域研究(研究領域提案型)

    研究期間:

    2012年06月
    -
    2017年03月
     

    長井 志江, 田中 文英, 尾形 哲也, 吉川 雄一郎, 西出 俊, 浅田 稔

     概要を見る

    社会的認知発達の基盤として「感覚・運動信号の予測学習」に基づく理論を提案し,計算論的視点から認知発達の構成的理解と発達障害者支援システムの開発に取り組んだ.乳幼児-養育者相互作用の解析では,乳幼児の身体的・社会的随伴性の発達と,それを促進する養育者の行動変容を明らかにした.計算論的発達モデルの研究では,多様な認知機能が予測学習に基づき獲得されること,モデルのパラメータ変動が発達障害を生み出すことを明らかにした.発達障害者の理解・支援システムの開発では,社会性の問題を引き起こす感覚・運動の非定型性を定量化・再現するシステムを実現した.これらの成果は認知発達原理の解明に大きく貢献している.

  • 長期インタラクション創発を可能とする知能化空間の設計論

    科学技術振興機構  さきがけ

    研究期間:

    2009年10月
    -
    2015年03月
     

  • ロボットの能動知覚に基づく物体挙動予測モデルと道具使用

    日本学術振興会  科学研究費助成事業 基盤研究(B)

    研究期間:

    2009年
    -
    2012年
     

    尾形 哲也, 西出 俊

     概要を見る

    本研究課題では,ロボットによる多様な一般物体の認識と道具使用を目的として,身体を用いた能動的知覚行為による物体挙動予測のモデルを実ロボットに具現化した.具体的には再帰結合型神経回路モデルと人間型ロボットを利用した,(1) 物体挙動に伴う音響信号の予測と分類,(2) 予測信頼性に基づく身体と物体の選択学習,さらに(3) 道具身体化モデルを構築した.本研究の成果は,工学的応用のみならず認知科学等の学術的な研究への展開も期待できる.

  • 音環境理解に基づくロボット聴覚の構築

    日本学術振興会  科学研究費助成事業 基盤研究(S)

    研究期間:

    2007年
    -
    2011年
     

    奥乃 博, 尾形 哲也, 駒谷 和範, 高橋 徹, 白松 俊, 中臺 一博, 北原 鉄朗, 糸山 克寿, 浅野 太, 浅野 太

     概要を見る

    音環境理解の主要機能である混合音の音源定位・分離・認識技術を開発し,ロボット聴覚ソフトウエアHARKとして公開し,国内外で複数の講習会を実施した. HARKを応用し,複数話者同時発話を認識する聖徳太子ロボット,ユーザの割込発話を許容する対話処理などを開発し,その有効性を実証した.さらに,多重奏音楽演奏から書くパート演奏を聞き分ける技術,実時間楽譜追跡機能を開発し,人と共演をする音楽ロボットなどに応用した。

  • 記号過程を内包した動的適応システムの設計論

    日本学術振興会  科学研究費助成事業 学術創成研究費

    研究期間:

    2007年
    -
    2011年
     

    椹木 哲夫, 土屋 和雄, 門内 輝行, 冨田 直秀, 横小路 泰義, 尾形 哲也, 青柳 富誌生, 水山 元, 中西 弘明, 堀口 由貴男, 青井 伸也, 谷口 忠大

     概要を見る

    複雑なシステムの中におかれたヒトや生体は,自らを取り巻くところの環境や社会を能動的に意味づけ,価値づけ,自らの棲む世界として秩序化していくことができる.本研究課題では,このような自律的主体の「多様性の生成と選択」の機構を「記号過程」に求め,記号の生成・利用のダイナミズムの観点から,生体細胞から環境適応機械(ロボット),社会組織に亘る様々なレベルにおける適応システムの同型性を見いだし,個々のシステム要素が外部・内部の物理的環境との相互作用を介して機能が形成される一般的過程について追究した. 5カ年の成果により,目的をもって生きる存在としての自律的な主体(人,ロボット,細胞)が, 他者主体を含む環境との相互作用を通して,意味の世界を創出して伝達する仕組み(記号過程)を解明し,システムが人を育て,人がシステムを育てる相互主導性を担保できるシステムの設計論を確立することができた。

  • オープンエンドな人間とロボットの協調における音声インタラクション創発に関する研究

    日本学術振興会  科学研究費助成事業 若手研究(A)

    研究期間:

    2005年
    -
    2007年
     

    尾形 哲也

     概要を見る

    本課題では,ロボットと人間との相互協調系について"力学系の視点からの解析"と"人間とのインタラクション実験"によって,その特徴を明らかにすることを目的としている.
    人間同士の相互協調を考える際,互いの行為モデルを獲得することは不可欠である.しかし人間は複雑な行為モデルを有するため,その完全なモデルを獲得する事は不可能であり,現実には可能な限り早く"有効な近似モデル"を獲得することが重要となる.平成19年度はこれまでの研究をベースに,「他者」モデル獲得をテーマとして研究を展開した.具体的に本研究では「幼児が外部環境や協調者(親)を理解する際,自己身体の順逆モデルを変換し,再利用することで学習コストを軽減している」という仮説を立て,ロボットを幼児に見立てた物体操作模倣(Emulation)発達過程モデルを提案した.
    提案モデルは以下の5つの学習段階を持つ.(step 1)ロボットによる自己身体・物体操作のバブリングと再起結合神経回路モデルを用いた順逆モデル獲得,(step 2)協調者とロボットにおける同一物体操作によるインタラクション(三項関係),(step 3)step 2における協調者によるロボットの模倣,(step 4)混合エキスパート神経回路モデルによる入出力(視野)変換モデルの自己組織化,(step 5)ロボットによる協調者行為予測のための変換モデル選択と模倣.
    実験の結果,ロボットは協調者予測のための変換モデルを少数回インタラクションで獲得し,これを用い他者(協調者)の視点での物体操作模倣が可能であることが分かった.さらにインタラクション中に未出現の物体操作の模倣,物体操作中の静止画像からの全体動作連想などが可能であることを確認した.これらを模倣に関する知見(Piaget,1996やRizzolatti,1998など)と比較し,その妥当性を議論した.今後はより高次の模倣(MimicやImitation)への拡張と人間とロボットの相互適応系への応用に取り組んでいく.

  • 環境音響を利用したロボットの動作生成

    日本学術振興会  科学研究費助成事業 萌芽研究

    研究期間:

    2005年
    -
    2006年
     

    尾形 哲也

     概要を見る

    本研究では"異種の感覚モダリティのマッピング"を工学的手法により実現することを目的とする.特に視覚と聴覚の変換に着目し,音響信号を表現するロボットモーションや音響モデルを再生する声道モデルダイナミクスを生成する.本技術は,多自由度ロボットのダンスパターンの自動生成や,環境音の擬音語化などに応用可能と期待される.さらに"共感覚"などの心理学の知見との関連に関する議論も期待できる.
    我々は平成17年度に小型ロボットKeepon(NICT開発)を用いた動作と音響信号の変換モデルを提案した.しかしその音響信号生成は白色信号をベースとした単純なモデルを用いていたため,きわめて多様性に乏しいものであった.そこで18年度は多様な音響信号生成,さらに幼児の発音過程のモデル化を目的として,平成17年度提案したモダリティ変換モデルに,声道モデル(Maedaモデル,1990)を適用した音声模倣モデルを構築した.具体的には声道モデルによる音声バブリングを人工神経回路で学習し,その体験をもとに入力音声を再生するモデルである.
    人間の実発話(母音)模倣をターゲットとして提案モデルの特性分析実験を行った.その結果,2母音の発話入力がバブリング済である場合,人間が弁別可能な高い品質で音声模倣を行えることを確認した.3母音の音声模倣では,提案モデルの汎化能力によりバブリングしていない音声パターンにおける模倣も実現した.また音響信号のみを学習した神経回路モデルと,音響信号と声道動作を学習した神経回路モデルとを比較し「音声知覚の運動理論」を支持する結果が得られた.

  • 人間とロボットの相互学習系におけるインタラクションの創発・発達に関する研究

    日本学術振興会  科学研究費助成事業 特定領域研究

    研究期間:

    2004年
    -
    2005年
     

    尾形 哲也