Kazuhiro Nakadai

J-GLOBAL         Last updated: Aug 8, 2019 at 14:34
 
Avatar
Name
Kazuhiro Nakadai
Nickname
nakadai
URL
http://www.ra.sc.e.titech.ac.jp/
Affiliation
Tokyo Institute of Technology
Section
School of Engineering
Job title
Specially-appointed professor
Degree
Ph. D.(The Univ. of Tokyo)
Other affiliation
Honda Research Institute Japan

Profile

Kazuhiro Nakadai received a B.E. in electrical engineering in 1993, an M.E. in information engineering in 1995, and a Ph.D. in electrical engineering in 2003 from the University of Tokyo. He worked with Nippon Telegraph and Telephone for four years as a system engineer from 1995 to 1999. After that, he was worked on the Kitano Symbiotic Systems Project, ERATO, JST as a researcher from 1999 to 2003. Currently he is a principal researcher for Honda Research Institute Japan, Co., Ltd. He has had a concurrent position at Tokyo Institute of Technology, as a visiting associate professor from 2006 to 2010, a visiting professor from 2011 to 2017, and a specially-appointed professor from July, 2017. He also had a concurrent position as a guest professor at Waseda University from 2011 to 2018. His research interests include AI, robotics, signal processing, computational auditory scene analysis, multi-modal integration and robot audition. He has been an executive board member for JSAI from 2015 to 2016, and for RSJ from 2017 to 2018. He is also a member of IPSJ, ASJ, HIS, ISCA, ACM and IEEE.

Research Areas

 
 

Academic & Professional Experience

 
Apr 2016
 - 
Today
Specially-appointed Professor, School of Engineering, Department of Systems and Control Engineering, Tokyo Institute of Technology
 
May 2003
 - 
Today
Principal Scientist, Honda Research Inst. Japan Co., Ltd.
 
Apr 2011
 - 
Mar 2018
Guest Professor, School of Creative Science and Engineering, Waseda University
 
Apr 2006
 - 
Mar 2016
Adjunct Associate Professor -> Adjunct Professor (2012), Graduate School of Information Science and Engineering, Tokyo Institute of Technology
 
Jul 1999
 - 
Apr 2003
Researcher, JST ERATO Kitano Symbiotic Systems Project
 

Education

 
Apr 1993
 - 
Mar 1995
Information Engineering, Graduate School of Engineering, The University of Tokyo
 
Apr 1991
 - 
Mar 1993
Department of Electrical and Electronics Engineering, Faculty of Engineering, The University of Tokyo
 
Apr 1989
 - 
Mar 1991
Natural Sciences I, School of Arts and Sciences, The University of Tokyo
 

Awards & Honors

 
Oct 2018
best generation award of innovation program, Ministry of Internal Affairs and Communications
 
Sep 2018
The 36th Annual Conference of the Robotics Society of Japan (RSJ 2018) International Session BEST PAPER AWARD, The Robotics Society of Japan
 
Sep 2017
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2017) Best Paper Award Finalist on Safety, Security, and Rescue Robotics (in memory of Motohiro Kisoi), IEEE
 
Sep 2016
Best Paper Award, Advanced Robotics, The Robotics Society of Japan
 
Jun 2016
Incentive Award, JSAI
 

Published Papers

 
Kenzo Nonami,Kotaro Hoshiba,Kazuhiro Nakadai,Makoto Kumon,Hiroshi G. Okuno,Yasutada Tanabe,Koichi Yonezawa,Hiroshi Tokutake,Satoshi Suzuki,Kohei Yamaguchi,Shigeru Sunada,Takeshi Takaki,Toshiyuki Nakata,Ryusuke Noda,Hao Liu 0016,Satoshi Tadokoro
Disaster Robotics - Results from the ImPACT Tough Robotics Challenge   77-142   2019   [Refereed]
Makoto Kumon,Kai Washizaki,Kazuhiro Nakadai
IEEE/SICE International Symposium on System Integration, SII 2019, Paris, France, January 14-16, 2019   313-318   2019   [Refereed]
Nelson Yalta,Shinji Watanabe,Takaaki Hori,Kazuhiro Nakadai,Tetsuya Ogata
CoRR   abs/1811.02735    2018   [Refereed]
Nelson Yalta,Shinji Watanabe,Kazuhiro Nakadai,Tetsuya Ogata
CoRR   abs/1807.01126    2018   [Refereed]
Ryosuke Taniguchi,Kotaro Hoshiba,Katsutoshi Itoyama,Kenji Nishida,Kazuhiro Nakadai
27th IEEE International Symposium on Robot and Human Interactive Communication, RO-MAN 2018, Nanjing, China, August 27-31, 2018   955-960   2018   [Refereed]
Agathe Balayn,Heike Brock,Kazuhiro Nakadai
27th IEEE International Symposium on Robot and Human Interactive Communication, RO-MAN 2018, Nanjing, China, August 27-31, 2018   370-377   2018   [Refereed]
Heike Brock,Shigeaki Nishina,Kazuhiro Nakadai
Proceedings of the 18th International Conference on Intelligent Virtual Agents, IVA 2018, Sydney, NSW, Australia, November 05-08, 2018   331-332   2018   [Refereed]
Daniel Gabriel,Ryosuke Kojima,Kotaro Hoshiba,Katsutoshi Itoyama,Kenji Nishida,Kazuhiro Nakadai
IEEE/SICE International Symposium on System Integration, SII 2019, Paris, France, January 14-16, 2019   199-204   2019   [Refereed]
Heike Brock,Kazuhiro Nakadai
Proceedings of the Eleventh International Conference on Language Resources and Evaluation, LREC 2018, Miyazaki, Japan, May 7-12, 2018.      2018   [Refereed]
Shinji Sumitani,R. Suzuki,Naoaki Chiba,Shiho Matsubayashi,Takaya Arita,Kazuhiro Nakadai,Hiroshi Gitchang Okuno
IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2019, Brighton, United Kingdom, May 12-17, 2019   8246-8250   2019   [Refereed]
Daniel Gabriel,Ryosuke Kojima,Kotaro Hoshiba,Katsutoshi Itoyama,Kenji Nishida,Kazuhiro Nakadai
Advanced Robotics   33(7-8) 403-414   2019   [Refereed]
Ryu Takeda,Kazuhiro Nakadai,Kazunori Komatani
2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018, Madrid, Spain, October 1-5, 2018   2503-2510   2018   [Refereed]
Ryosuke Kojima,Osamu Sugiyama,Kotaro Hoshiba,Reiji Suzuki,Kazuhiro Nakadai
2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018, Madrid, Spain, October 1-5, 2018   2497-2502   2018   [Refereed]
Shinji Sumitani,Reiji Suzuki,Shiho Matsubayashi,Takaya Arita,Kazuhiro Nakadai,Hiroshi G. Okuno
2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018, Madrid, Spain, October 1-5, 2018   2485-2490   2018   [Refereed]
Array,Katsutoshi Itoyama,Array,Satoshi Tadokoro,Array,Kazuyoshi Yoshii,Tatsuya Kawahara,Array
IEEE/ACM Trans. Audio, Speech & Language Processing   26(2) 215-230   2018   [Refereed]
Kazuhiro Nakadai,Emilia I. Barakova,Michita Imai,Tetsunari Inamura
Advanced Robotics   33(7-8) 307-308   2019   [Refereed]
尾崎翔, 浅野太, 中臺一博
電子情報通信学会論文誌 A(Web)   J101-A(6) 137‐149 (WEB ONLY)   Jun 2018
Suzuki R, Matsubayashi S, Saito F, Murate T, Masuda T, Yamamoto K, Kojima R, Nakadai K, Okuno HG
Ecology and evolution   8(1) 812-825   Jan 2018   [Refereed]
Kotaro Hoshiba,Kazuhiro Nakadai,Makoto Kumon,Hiroshi G. Okuno
JRM   30(3) 426-435   2018   [Refereed]
Hiroshi G. Okuno,Kazuhiro Nakadai
JRM   29(1) 15   2017   [Refereed]
Ryu Takeda,Kazuhiro Nakadai,Kazunori Komatani
Interspeech 2017, 18th Annual Conference of the International Speech Communication Association, Stockholm, Sweden, August 20-24, 2017   1636-1640   2017   [Refereed]
Kazuhiro Nakadai,Makoto Kumon,Hiroshi G. Okuno,Kotaro Hoshiba,Mizuho Wakabayashi,Kai Washizaki,Takahiro Ishiki,Daniel Gabriel,Yoshiaki Bando,Takayuki Morito,Ryosuke Kojima,Osamu Sugiyama
2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017, Vancouver, BC, Canada, September 24-28, 2017   5985-5990   2017   [Refereed]
Ryu Takeda,Kazuhiro Nakadai,Kazunori Komatani
Computer Speech & Language   46 461-480   2017   [Refereed]
Ryosuke Kojima,Osamu Sugiyama,Kotaro Hoshiba,Reiji Suzuki,Kazuhiro Nakadai
2017 IEEE International Conference on Data Science and Advanced Analytics, DSAA 2017, Tokyo, Japan, October 19-21, 2017   395-404   2017   [Refereed]
Kotaro Hoshiba,Kai Washizaki,Mizuho Wakabayashi,Takahiro Ishiki,Array,Yoshiaki Bando,Daniel Gabriel,Kazuhiro Nakadai,Hiroshi G. Okuno
Sensors   17(11) 2535   Nov 2017   [Refereed]
Lana Sinapayen,Keisuke Nakamura,Kazuhiro Nakadai,Hiroki Takahashi,T. Kinoshita
Advanced Robotics   31(12) 624-633   2017   [Refereed]
Ryosuke Kojima,Osamu Sugiyama,Kotaro Hoshiba,Kazuhiro Nakadai,Reiji Suzuki,Charles E. Taylor
JRM   29(1) 236-246   2017   [Refereed]
Shiho Matsubayashi,Reiji Suzuki,Fumiyuki Saito,Tatsuyoshi Murate,Tomohisa Masuda,Koichi Yamamoto,Ryosuke Kojima,Kazuhiro Nakadai,Hiroshi G. Okuno
JRM   29(1) 224-235   2017   [Refereed]
Reiji Suzuki,Shiho Matsubayashi,Richard W. Hedley,Kazuhiro Nakadai,Hiroshi G. Okuno
JRM   29(1) 213-223   2017   [Refereed]
Osamu Sugiyama,Satoshi Uemura,Akihide Nagamine,Ryosuke Kojima,Keisuke Nakamura,Kazuhiro Nakadai
JRM   29(1) 188-197   2017   [Refereed]
Takuma Ohata,Keisuke Nakamura,Akihide Nagamine,Takeshi Mizumoto,Takayuki Ishizaki,Ryosuke Kojima,Osamu Sugiyama,Kazuhiro Nakadai
JRM   29(1) 177-187   2017   [Refereed]
Kotaro Hoshiba,Osamu Sugiyama,Akihide Nagamine,Ryosuke Kojima,Makoto Kumon,Kazuhiro Nakadai
JRM   29(1) 154-167   2017   [Refereed]
Kazuhiro Nakadai,Taiki Tezuka,Takami Yoshida
JRM   29(1) 114-124   2017   [Refereed]
Kazuhiro Nakadai,Tomoaki Koiwa
JRM   29(1) 105-113   2017   [Refereed]
Nelson Yalta,Kazuhiro Nakadai,Array
JRM   29(1) 37-48   2017   [Refereed]
Kazuhiro Nakadai,Array,Takeshi Mizumoto
JRM   29(1) 16-25   2017   [Refereed]
Sakata Naoto, Murakami Tetsuro, Nakajima Hirofumi, Nakadai Kazuhiro
THE JOURNAL OF THE ACOUSTICAL SOCIETY OF JAPAN   72(12) 739-748   2016
<p>風雑音は一般的に非定常な雑音であり,信号の波形レベルでの相関をもとにした処理についてはあまり行われていない。本論文では2チャンネルを近接させたマイクロホンを用いて,各チャンネルで相関のある風雑音の収録を行い,相関の分析・風雑音の低減の二つの実験を行った。振幅・パワー・複素信号のそれぞれについてコヒーレンス関数により相関を分析した結果,どの項目についても125Hz以下で0.3~0.8の相関が確認された。その相関を利用して2種類の線形ビームフォーマにより風雑音の低減を行い,125Hz以下...
石井 健太郎, 谷口 祐司, 大澤 博隆, 中臺 一博, 今井 倫太
情報処理学会論文誌   54(4) 1413-1421   Apr 2013
本論文では,仮想的な身体を持つアバタを投影する遠隔コミュニケーションシステムPROT AVATARにおけるアバタ操作手法に関する実験をもとに,得られた知見について議論する.PROT AVATARによるコミュニケーションでは,アバタの操作者の映像を遠隔地に投影するため,表情により感情を伝えることができる.さらに,アバタの操作者にとっては明確ではない,アバタの投影に適切な位置をシステムが自動で計算するため,アバタの操作者は投影位置を考えることなく遠隔の環境内を指し示すことができる.しかし,アバ...
NAKAMURA Keisuke, NAKADAI Kazuhiro, ASANO Futoshi, NAKAJIMA Hirofumi, INCE Gokhan
計測自動制御学会論文集 = Transactions of the Society of Instrument and Control Engineers   48(6) 349-358   Jun 2012
Takeda R, Nakadai K, Takahashi T, Komatani K, Ogata T, Okuno HG
Neural computation   24(1) 234-272   Jan 2012   [Refereed]
YOSHIDA Takami, NAKADAI Kazuhiro, OKUNO Hiroshi G.
Journal of the Robotics Society of Japan   28(8) 970-977   Oct 2010
Yoshida Takami, Nakadai Kazuhiro, Okuno Hiroshi G.
Journal of the Robotics Society of Japan   28(8) 970-977   2010
Noise-robust Automatic Speech Recognition (ASR) is essential for robots which are expected to communicate with human in a daily environment. In such an environment, Voice Activity Detection (VAD) performance becomes poor, and ASR performance deter...
FUJIMURA Ryota, GUO BIN, OHMURA Ren, NAKADAI Kazuhiro, IMAI Michita
Journal of Japan Society for Fuzzy Theory and Intelligent Informatics   21(5) 701-712   Oct 2009
In this paper, we describe a movable projection avatar system named &ldquo;Remy&rdquo;. Remy aims to support communication that share other party's actual environment from a remote environment. There are three issues in a lot of existing studies i...
NAKAJIMA Hirofumi, NAKADAI Kazuhiro, HASEGAWA Yuji, TSUJINO Hiroshi
Journal of the Robotics Society of Japan   27(7) 774-781   Sep 2009
This paper describes a novel sound source separation method for a robot that needs to cope with dynamically changing noises in the real world. A sound source separation method, Geometric Source Separation (GSS), is promising because it has high se...
TAKEDA Ryu, NAKADAI Kazuhiro, TAKAHASHI Toru, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
Journal of the Robotics Society of Japan   27(7) 782-792   Sep 2009
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 ...
MURATA Kazumasa, NAKADAI Kazuhiro, TAKEDA Ryu, OKUNO Hiroshi G., HASEGAWA Yuji, TSUJINO Hiroshi
Journal of the Robotics Society of Japan   27(7) 793-801   Sep 2009
Human-robot interaction through music in real environments is essential for robots, because such a robot makes people enjoyable. To deal with real music signals by using robot's own ears, we propose a beat-tracking algorithm for a robot based on s...
OKUNO Hiroshi G., NAKADAI Kazuhiro, OHTSUKA Takuma
IPSJ Magazine   50(8) 729-734   Aug 2009
音楽のリズムに合わせて振舞う音楽ロボットを目標に据えると, 音楽情報処理の課題が見えてくる.
TAKEDA Ryu, NAKADAI Kazuhiro, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
Journal of the Robotics Society of Japan   26(6) 529-536   Aug 2008
This paper describes a new adaptive filter algorithm based on independent component analysis (ICA) for enhancing a target sound and for suppressing other interference sounds that are known. The technique can provide barge-in capable robot audition...
Nakajima Hirofumi, Nakadai Kazuhiro, Hasegawa Yuji, Tsujino Hiroshi
NEW FRONTIERS IN ARTIFICIAL INTELLIGENCE   4914 47-53   2008   [Refereed]
2008 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING, VOLS 1-12   149-152   2008   [Refereed]
A robot referee for rock-paper-scissors sound games
2008 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION, VOLS 1-9   3469-+   2008   [Refereed]
Computational auditory scene analysis and its application to robot audition
2008 HANDS-FREE SPEECH COMMUNICATION AND MICROPHONE ARRAYS   125-+   2008   [Refereed]
Nakajima Hirofumi, Nakadai Kazuhiro, Hasegawa Yuuji, Tsujino Hiroshi
2008 IEEE/RSJ INTERNATIONAL CONFERENCE ON ROBOTS AND INTELLIGENT SYSTEMS, VOLS 1-3, CONFERENCE PROCEEDINGS   2165-2171   2008   [Refereed]
Murata Kazumasa, Nakadai Kazuhiro, Yoshii Kazuyoshi, Takeda Ryu, Torii Toyotaka, Okuno Hiroshi G., Hasegawa Yuji, Tsujino Hiroshi
2008 IEEE/RSJ INTERNATIONAL CONFERENCE ON ROBOTS AND INTELLIGENT SYSTEMS, VOLS 1-3, CONFERENCE PROCEEDINGS   2459-+   2008   [Refereed]
NISHIMURA Yoshitaka, ISHIZUKA Mitsuru, NAKADAI Kazuhiro, NAKANO Mikio, TSUJINO Hiroshi
Journal of the Robotics Society of Japan   25(8) 1189-1198   Nov 2007
Automatic speech recognition (ASR) is essential for human-humanoid communication. One of the main problems with ASR by a humanoid is that it is inevitably generates motor noises. These noises are easily captured by the humanoid's microphones becau...
NAKADAI Kazuhiro, NAKAJIMA Hirofumi, MURASE Masamitsu, OKUNO Hiroshi G., HASEGAWA Yuji, TSUJINO Hiroshi
Journal of the Robotics Society of Japan   25(6) 979-989   Sep 2007
Real-time and robust sound source tracking is an important function for a robot operating in a daily environment, because the robot should recognize where a sound event such as speech, music and other environmental sounds originates from. This pap...
NAKADAI Kazuhiro
計測と制御 = Journal of the Society of Instrument and Control Engineers   46(6) 427-433   Jun 2007
KANDA NAOYUKI, KOMATANI KAZUNORI, NAKANO MIKIO, NAKADAI KAZUHIRO, TSUJINO HIROSHI, OGATA TETSUYA, OKUNO HIROSHI G.
IPSJ journal   48(5) 1980-1989   May 2007
We have developed a robust domain selection method using dialogue history in multi-domain spoken dialogue systems. We define domain selection as a classifying problem among (I) the domain in the previous turn, (II) the domain in which N-best speec...
KUROTAKI Shunsuke, SUZUKI Noriaki, NAKADAI Kazuhiro, OKUNO Hiroshi G., AMANO Hideharu
The IEICE transactions on information and systems   90(3) 897-907   Mar 2007
近年,人間と共生するロボットが多数登場してきている.これらのロボットが人間と言語を用いたインタラクションを行うためには音声認識が必要となるが,従来の音声認識手法は単一音源を対象としているため,複数人の同時発話や周囲に雑音がある環境では著しく認識精度が低下してしまうという問題がある.よって,実環境での音声認識にはその前処理として,混合音から注目する音声信号のみを抽出する音源分離処理が不可欠となる。実時間で音源分離を行うためには多大な計算コストを要する一方で,自律型のロボットは消費電力やシステ...
YAMAMOTO Shunichi, VALIN Jean-Marc, NAKADAI Kazuhiro, NAKANO Mikio, TSUJINO Hiroshi, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
Journal of the Robotics Society of Japan   25(1) 92-102   Jan 2007
Our goal is to realize a humanoid robot that has the capabilities of recognizing simultaneous speech. A humanoid robot under real-world environments usually hears a mixture of sounds, and thus three capabilities are essential for robot audition; s...
NAKADAI Kazuhiro
Journal of The Society of Instrument and Control Engineers   46(6) 427-433   2007
The design of phoneme grouping for coarse phoneme recognition
Nakadai Kazuhiro, Sumiya Ryota, Nakano Mikio, Ichige Koichi, Hirose Yasuo, Tsujino Hiroshi
NEW TRENDS IN APPLIED ARTIFICIAL INTELLIGENCE, PROCEEDINGS   4570 905-+   2007   [Refereed]
A biped robot that keeps steps in time with musical beats while listening to music with its own ears
2007 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-9   1749-+   2007   [Refereed]
2007 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-9   1757-1762   2007   [Refereed]
Exploiting known sound source signals to improve ICA-based robot audition in speech separation and recognition
2007 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-9   1763-+   2007   [Refereed]
YAMAMOTO Shunichi, NAKADAI Kazuhiro, NAKANO Mikio, TSUJINO Hiroshi, VALIN Jean-Marc, TAKEDA Ryu, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
Human interface   8(2) 203-212   May 2006
NAKADAI Kazuhiro, TSUJINO Hiroshi
Human interface   8(2) 213-221   May 2006
YAMAMOTO Shunichi, NAKADAI Kazuhiro, TSUJINO Hiroshi, OKUNO Hiroshi G.
Journal of the Robotics Society of Japan   23(6) 743-751   Sep 2005
Robot audition is a critical technology in creating an intelligent robot operating in daily environments. To realize such a robot audition system, we have designed a missing feature theory based interface between sound source separation and automa...
NAKADAI Kazuhiro, OKUNO Hiroshi G., KITANO Hiroaki
Transactions of the Japanese Society for Artificial Intelligence   18 104-113   Nov 2003
In this paper, we present an active audition system which is implemented on the humanoid robot "SIG the humanoid". The audition system for highly intelligent humanoids localizes sound sources and recognizes auditory events in the auditory s...
NAKADAI Kazuhiro, HIDAI Ken-ichi, MIZOGUCHI Hiroshi, OKUNO Hiroshi, KITANO Hiroaki
Journal of the Robotics Society of Japan   21(5) 517-525   Jul 2003
This paper describes a real-time human tracking system by audio-visual integrtation for the humanoid SIG. An essential idea for real-time and robust tracking is hierarchical integration of multi-modal information. The system creates three k...
中臺 一博, 奥乃 博, 北野 宏明
人工知能学会論文誌(Transactions of the Japanese Society for Artificial Intelligence)   18(2) 104-113   Jan 2003
In this paper, we present an active audition system which is implemented on the humanoid robot "SIG the humanoid". The audition system for highly intelligent humanoids localizes sound sources and recognizes auditory events in the auditory scene. A...
KINOSHITA Tomoyoshi, HANDA Ibuki, MUTO Makoto, SAKAI Shuichi, TANAKA Hidehiko
The transactions of the Institute of Electronics, Information and Communication Engineers. D-II   85(3) 373-381   Mar 2002
音響信号により外界の事象を理解する聴覚的情景分析に関して,従来多くの研究がなされてきた.特に対象を音楽に絞った場合,自動採譜等の実現を目指した研究例がいくつかある.しかしながら,従来の処理では各時点における局所的な処理に終始するものが多く,時間方向の処理を進めた例であっても,その対象は時間的に近接した範囲にとどまっていたため,処理性能に限界があった.本論文では,それを改善することを目的として,音響ストリームの知覚的な階層構造に着目し,より大局的な範囲での処理を用いて,楽曲からパートに相当す...
KINOSHITA Tomoyoshi, SAKAI Shuichi, TANAKA Hidehiko
The Transactions of the Institute of Electronics,Information and Communication Engineers.   83(4) 1073-1081   Apr 2000
音響信号により外界の事象を理解する聴覚的情景分析に関して, 従来多くの研究がなされてきた.特に対象を音楽に絞った場合, 自動採譜等の実現を目指し, いくつかの研究例がある.その一つとして, 筆者らはこれまでに音楽音響信号を対象とした聴覚的情景分析の処理モデルOPTIMAを提案し, その実験システムを構築した.しかしながら, その認識精度は実用上十分とはいえず, その改善が課題となっている.本論文では, 従来の処理の問題点である周波数成分の重なりに対する脆弱性を改善するための新たな処理を提案...
KASHINO Kunio, NAKADAI Kazuhiro, KINOSHITA Tomoyoshi, TANAKA Hidehiko
The Transactions of the Institute of Electronics,Information and Communication Engineers.   79(11) 1762-1770   Nov 1996
音楽演奏の音響信号を対象として演奏情報を認識する試みとしては,従来自動採譜の研究が行われているが,複数種類の楽器音を含む音楽演奏を対象とする場合には,認識処理の有効性は極めて限られていた.そこで本論文では,複数種類の楽器音を含む音楽演奏の認識を音楽情景分析の問題としてとらえ,その解決を図る.ここで音楽情景分析とは,音楽演奏の音響信号から,単音や和音などの音楽演奏情報を記号表現として抽出することを指す.本論文ではまず,音楽情景分析を実現する上では情報統合の技術が不可欠であるとの認識から,ベイ...
KASHINO Kunio, KINOSHITA Tomoyoshi, NAKADAI Kazuhiro, TANAKA Hidehiko
The transactions of the Institute of Electronics, Information and Communication Engineers   79(11) 1762-1770   1996
我々は,複数種類の楽器音を含む音楽演奏を対象とした音楽認識を,音楽情景分析の問題としてとらえ研究を行っている.ここで音楽情景分析とは,音楽演奏の音響信号から,単音や和音などの音楽演奏情報を記号表現として抽出することを指す.我々は先に,ベイジアンネットワークによる情報統合の機構を備えた音楽情景分析の処理モデルOPTIMAを提案した.本論文では,OPTIMAにおける処理のうち,特に和音の認識に的を絞って,情報統合機構の有効性を調べた.その結果,サンプル曲を用いた評価実験において,ボトムアップ処...

Misc

 
Nakadai Kazuhiro
SYSTEMS, CONTROL AND INFORMATION   62(2) 42-49   Aug 2018
中臺一博, 中臺一博
日本音響学会誌   74(7) 394‐400   Jul 2018
山本 俊一, 住田 直亮, 中臺 一博
Honda R&D technical review   29(2) 110-117   Oct 2017
中臺 一博
映像情報メディア学会誌 = The journal of the Institute of Image Information and Television Engineers   71(5) 647-653   Sep 2017
中臺 一博, 小林 一郎, 和泉 潔
人工知能 : 人工知能学会誌 : journal of the Japanese Society for Artificial Intelligence   32(2) 297-304   Mar 2017
bando Yoshiaki, Ambe Yuichi, Itoyama Katsutoshi, Konyo Masashi, Tadokoro Satoshi, Nakadai Kazuhiro, Yoshii Kazuyoshi, G. Okuno Hiroshi
The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)   2017(0) 1P2-P05   2017
<p>This paper presents a real-time human-voice enhancement method for a hose-shaped rescue robot based on multi-channel low-rank sparse decomposition. Although microphone arrays equipped on hose-shaped robots are crucial for finding victims under ...
HOSHIBA Kotaro, WASHIZAKI Kai, WAKABAYASHI Mizuho, KUMON Makoto, NAKADAI Kazuhiro
The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)   2017(0) 1P1-R05   2017
<p>Sound source localization using a microphone array embedded on an unmanned aerial vehicle has been studied to detect and localize people who need help in a disaster-stricken area. Because such sound source localization should work in outdoor en...
和泉 潔, 中臺 一博, 栗原 聡
人工知能 : 人工知能学会誌 : journal of the Japanese Society for Artificial Intelligence   31(4) 531-549,530   Jul 2016
蓮本 諒介, 小山 大幾, 水本 武志, 中村 圭佑, 中臺 一博, 今井 倫太
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   116(461) 19-22   Feb 2017
蓮本 諒介, 小山 大幾, 水本 武志, 中村 圭佑, 中臺 一博, 今井 倫太
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   116(462) 19-22   Feb 2017
Nakadai Kazuhiro, Nakamura Keisuke, Tezuka Taiki
Proceedings of the IEICE General Conference   2014(2) "SS-18"-"SS-19"   Mar 2014
Nakadai Kazuhiro, Okuno Hiroshi G., Mizumoto Takeshi, Nakamura Keisuke
シミュレーション = Journal of the Japan Society for Simulation Technology   35(1) 32-38   Mar 2016
Nakadai Kazuhiro, Nakamura Keisuke
THE JOURNAL OF THE ACOUSTICAL SOCIETY OF JAPAN   70(7) 397-402   2014
Jean-Marc Valin, Shun'ichi Yamamoto, Jean Rouat, Francois Michaud, Kazuhiro Nakadai, Hiroshi G. Okuno
IEEE Transactions on Robotics, Vol. 23, No. 4, pp. 742-752, 2007      Feb 2016
This paper describes a system that gives a mobile robot the ability to
perform automatic speech recognition with simultaneous speakers. A microphone
array is used along with a real-time implementation of Geometric Source
Separation and a post-filt...
中島 弘史, 坂田 直人, 加科 優希, 中臺 一博
回路とシステムワークショップ論文集 Workshop on Circuits and Systems   28 208-213   Aug 2015
坂田 直人, 村上 哲郎, 中島 弘史, 中臺 一博
回路とシステムワークショップ論文集 Workshop on Circuits and Systems   28 359-364   Aug 2015
SAKATA Naoto, NAKAJIMA Hirofumi, NAKADAI Kazuhiro
IEICE technical report. Signal processing   114(474) 1-6   Mar 2015
In this report, wind-induced noise reduction in time domain was investigated using closely-aligned two microphones. A linear beamforming filter in frequency domain on the basis of time frame decomposition was applied to signals in time domain. The...
SAKATA Naoto, NAKAJIMA Hirofumi, NAKADAI Kazuhiro
Technical report of IEICE. EA   114(473) 1-6   Mar 2015
In this report, wind-induced noise reduction in time domain was investigated using closely-aligned two microphones. A linear beamforming filter in frequency domain on the basis of time frame decomposition was applied to signals in time domain. The...
SAKATA Naoto, NAKAJIMA Hirofumi, NAKADAI Kazuhiro
IEICE technical report. Speech   114(475) 1-6   Mar 2015
In this report, wind-induced noise reduction in time domain was investigated using closely-aligned two microphones. A linear beamforming filter in frequency domain on the basis of time frame decomposition was applied to signals in time domain. The...
TAKAHASHI Masaaki, OGATA Masa, IMAI Michita, NAKAMURA Keisuke, NAKADAI Kazuhiro
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   114(351) 1-5   Dec 2014
The study of the telepresence robot becomes popular as a communication tool in the remote place. However, there is a problem that the telepresence system can't precisely transfer the user's utterance because of not considering difference of sound ...
ラナシナパヤ, 中村圭佑, 中臺一博, 高橋秀幸, 木下哲男
第76回全国大会講演論文集   2014(1) 185-186   Mar 2014
We propose a novel approach to multicopter localization, using sound landmarks and one embedded microphone. This approach can benefit to multicopter localization in that it requires less computational power and smaller payloads than image-based ap...
Koike Kyotaro, Imai Michita, Nakamura Keisuke, Nakadai Kazuhiro
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   113(372) 1-6   Dec 2013
A telepresence robot is useful to deal with a situation where a user in a remote area has to control the robot to communicate with people. However, there exists some remaining issues that the target speech is contaminated with unnecessary speeches...
HAYAMIZU Akira, IMAI Michita, NAKAMURA Keisuke, NAKADAI Kazuhiro
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   113(372) 35-40   Dec 2013
The Lombard effect is the involuntary tendency of speakers to increase their vocal effort when speaking in loud noise to enhance the audibility of their voice. There is a problem in a telecommunication situation due to the Lombard effect, and woul...
OKUTANI Keita, YOSHIDA Takami, NAKAMURA Keisuke, NAKADA Kazuhiro
Journal of the Robotics Society of Japan   31(7) 676-683   Sep 2013
This paper addresses sound source localization using an aerial vehicle with a microphone array in an outdoor environment to realize outdoor auditory scene analysis. It, for instance, aims at finding distressed people in a disaster situation. In su...
Moon Seong-eun, Takagi Kentaro, Kamashima Tsutomu, Nakadai Kazuhiro, Otake Mihoko
The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)   2013(0) _2P1-P24_1-_2P1-P24_2   2013
This paper presents a sound source localization system that composes a wireless microphone array named Jellyfish-02 and robot audition software HARK. Jellyfish-02 surpasses existing microphone array in design and usability, because it has a cover ...
OKUNO Hiroshi G., NAKADAI Kazuhiro, MIZUMOTO Takeshi
The Journal of the Institute of Electronics, Information, and Communication Engineers   95(5) 401-404   May 2012
私たちが日常耳にする音は複数の音や背景雑音が混じった混合音である.実世界で音情報を活用するためには「聞き分ける」機能が不可欠である.聞き分けるセンサ技術は,インストルメンテーション(装置化)という観点から音を収録するデバイス(センサ)と収録音に対する処理ソフトウェアから構成される.本稿では,混合音のセンサ技術の動向を,ロボット聴覚とカエルの合唱の観測について解説を行う.混合音を聞き分けるという立場から,音源定位,音源分離,分離音認識に取り組むべきであると考え,音環境理解という研究を過去15...
SUGIYAMA Osamu, ITOYAMA Katsutoshi, NAKADAI Kazuhiro, OKUNO Hiroshi G.
電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   114(85) 23-26   Jun 2014
In this study we designed and developed the multidirectional sound source annotation tool with the robot audition software, HARK. With the rise of inexpensive microphone array products and the robot audition software called HARK, we can record and...
NAKAMURA Keisuke, NAKADAI Kazuhiro, ASANO Futoshi, NAKAJIMA Hirofumi, INCE G&ouml;khan
Transactions of the Society of Instrument and Control Engineers   48(6) 349-358   2012
Localization and tracking of humans are essential research topics in robotics. In particular, Sound Source Localization (SSL) has been of great interest. Despite the numerous reported methods, SSL in a real environment had mainly three issues; rob...
AngelicaLim, 中村圭佑, 中臺一博, 尾形哲也, 奥乃博
第73回全国大会講演論文集   2011(1) 309-310   Mar 2011
Is this person playing a violin or a flute? Classification of musical instrument performances is usually carried out using audio features such as spectral coefficients. We propose augmenting the typical audio feature set with visual features. We s...
G. OKUNO Hiroshi, NAKADAI Kazuhiro
The Journal of The Institute of Electrical Engineers of Japan   131(3) 159-163   Mar 2011
This article has no abstract.
Okuno Hiroshi G., Nakadai Kazuhiro, Takahashi Toru
Proceedings of the Society Conference of IEICE   2010 "SS-72"-"SS-73"   Aug 2010
奥乃 博, 中臺 一博
日本ロボット学会誌(Journal of the Robotics Society of Japan)   28(1) 6-9   Jan 2010
中臺 一博, 宮下 敬宏, 奥乃 博
日本ロボット学会誌(Journal of the Robotics Society of Japan)   28(1) 1-1   Jan 2010
NAKAMURA Keisuke, NAKADAI Kazuhiro, INCE Gokhan
IEICE technical report   111(32) 35-40   May 2011
Since scene recognition and robot perception have been of great interest, information integration has become a significant research topic in robotics. From the viewpoint of scalability and reusability, utilization of appropriate middleware is a ke...
NAKADAI Kazuhiro, HASEGAWA Yuji, SEKIGUCHI Tatsuhiko, TSUJINO Hiroshi
Journal of the Robotics Society of Japan   27(1) 6-9   Jan 2009
Nakadai Kazuhiro, Yamamoto Shunichi, Okuno Hiroshi G., Nakajima Hirofumi, Hasegawa Yuji, Tsujino Hiroshi
The Proceedings of JSME annual Conference on Robotics and Mechatronics (Robomec)   2008(0) _1P1-G13_1-_1P1-G13_4   2008
This paper describes an open source software system for robot audition called HARK (Honda Research Institute Japan Audition for Robots with Kyoto University). HARK consists of a lot of modules including multi-channel audio input, sound source loca...
NAKADAI Kazuhiro, OKUNO Hiroshi G.
IEICE technical report   110(401) 7-12   Jan 2011
This paper addresses robot audition, which realizes listening capabilities for robots using robot-embedded microphones. For robot audition, we propose real-time sound source separation and automatic speech recognition (ASR) techniques for dynamica...
Takeda Ryu, Nakadai Kazuhiro, Komatani Kazunori, Ogata Tetsuya, Okuno Hiroshi G.
情報科学技術フォーラム一般講演論文集   6(2) 261-262   Aug 2007
中臺 一博, 山本 俊一, 浅野 太
人工知能学会全国大会論文集   21 1-4   2007
山本 俊一, 中臺 一博, 辻野 広司
人工知能学会全国大会論文集   18 1-4   2004
OKUNO Hiroshi G., NAKADAI Kazuhiro
IPSJ Magazine   44(11) 1138-1144   Nov 2003
ロボットが家庭に入ってくるようになり, ロボットと人とのコミュニケーション, 特に, ロボットに装備されたマイクロフォンを用いたコミュニケーションや音による環境知覚がますます重要になってきている. 最近, ロボット自身の耳による聴覚機能がようやく活発になってきた. では, ロボットのための聴覚機能にはどのようなものが必要であろうか.
HARUBARA Takuya, NAKAJIMA Hirofumi, NAKADAI Kazuhiro, KANEDA Yutaka
IEICE technical report   110(131) 19-24   Jul 2010
This paper addresses a real-time sound source orientation estimation system using a 96ch microphone-array. We proposed a beam-forming method with estimation of sound source directivity, and reported orientation estimation of a speech source such a...
Okuno Hiroshi G., Nakadai Kazuhiro
THE JOURNAL OF THE ACOUSTICAL SOCIETY OF JAPAN   58(3) 205-210   2002
聴覚は人間にとって最も重要な感覚である。言語によるコミュニケーションが聴覚によって成立することは容易に理解されるが,「ヒトは聴覚によってのみ言語を獲得し,そこに文化が生まれ,継承される。書かれた言語は目によって伝承されるが,話す言葉は耳からしか得られない。話し言葉があって書く言葉が生まれる」ことを,多くの人が理解していないのは残念なことである(鈴木淳一,小林武夫共著『耳科学-難聴に挑む』(中公新書1598,2001))。
TAKEDA Ryu, NAKADAI Kazuhiro, TAKAHASHI Toru, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
全国大会講演論文集   72(0) 27-28   Mar 2010
TAKAHASHI Toru, NAKADAI Kazuhiro, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
全国大会講演論文集   72(0) 29-30   Mar 2010
NAKADAI Kazuhiro, MIYASHITA Takahiro, OKUNO Hiroshi G.
Journal of the Robotics Society of Japan   28(1)    Jan 2010
OKUNO Hiroshi G., NAKADAI Kazuhiro
Journal of the Robotics Society of Japan   28(1) 6-9   Jan 2010
NAKAJIMA Hirofumi, DAIGO Tohru, NAKADAI Kazuhiro, KANEDA Yutaka, HASEGAWA Yuji
IEICE technical report   109(136) 7-12   Jul 2009
This paper addresses a multi-stage processing mechanism that improves various dereverberation methods. In the mechanism, each stage is implemented as an intermediate processing module that connects the outputs of the modules on the previous stage ...
NAKAJIMA Hirofumi, DAIGO Tohru, NAKADAI Kazuhiro, KANEDA Yutaka, HASEGAWA Yuji
電子情報通信学会技術研究報告. EA, 応用音響   109(136) 7-12   Jul 2009
鈴木 淑正, 中島 弘史, 中臺 一博
聴覚研究会資料   39(4) 325-330   Jun 2009
SUZUKI Toshimasa, NAKAJIMA Hirofumi, NAKADAI Kazuhiro, ARAI Takahiro, HASEGAWA Yuji
IEICE technical report   109(100) 109-114   Jun 2009
Thanks to improvements in computer performance, numerical simulation based on wave acoustics works in practical time with off-the-shelf computers. Such a numerical simulation method accurately estimates a sound field when it is a simple and simula...
TAKAHASHI Toru, NAKADAI Kazuhiro, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
全国大会講演論文集   71(0) 35-36   Mar 2009
OTSUKA Takuma, MURATA Kazumasa, TAKEDA Ryu, NAKADAI Kazuhiro, TAKAHASHI Toru, OGATA Tetsuya, OKUNO Hiroshi G.
全国大会講演論文集   71(0) 243-244   Mar 2009
NISIMURA Ryuichi, NAKANO Teppei, KURIHARA Kazutaka, NAKADAI Kazuhiro, YOSHINO Takashi
IPSJ SIG Notes   2008(102) 55-60   Oct 2008
To induce developments of ASR applications, this panel discussion introduces actual case studies. We also indicate some problems of ASR application developments.
KIKUCHI Keiko, DAIGO Tohru, NAKAJIMA Hirofumi, NAKADAI Kazuhiro, HASEGAWA Yuji, KANEDA Yutaka
IEICE technical report   108(143) 13-18   Jul 2008
This paper addresses sound source orientation estimation using a 96ch microphone array. We proposed a beam-forming method with estimation of sound source directivity, and reported orientation estimation of a speech source such as a loudspeaker or ...
TAKEDA Ryu, NAKADAI Kazuhiro, KOMATANI Kazunori, OGATA Tetsuya, OKUNO Hiroshi G.
全国大会講演論文集   70(0) 135-136   Mar 2008
NAKAJIMA Hirofumi, NAKADAI Kazuhiro, HASEGAWA Yuji, TSUJINO Hiroshi
IEICE technical report   107(120) 19-24   Jun 2007
This paper describes a novel blind source separation (BSS) method. One of the most important factors in BSS performance is a step-size parameter to update a decomposition matrix which is generally used for extracting a target sound source. A fixed...
Nakadai Kazuhiro, Nakajima Hirofumi, Murase Masamitsu, Okuno Hiroshi G., Hasegawa Yuji, Tsujino Hiroshi
Proceedings of the IEICE General Conference   2007 "S-65"-"S-66"   Mar 2007
NAKADAI Kazuhiro
IEICE technical report   106(298) 19-26   Oct 2006
To realize natural human-robot interaction, we consider that a robot should have at least two functions, that is, real-world auditory scene analysis by a robot to understand the surrounding environments, and robot expression to send information to...
NAKADAI Kazuhiro
IEICE technical report   106(296) 19-26   Oct 2006
To realize natural human-robot interaction, we consider that a robot should have at least two functions, that is, real-world auditory scene analysis by a robot to understand the surrounding environments, and robot expression to send information to...
NAKADAI Kazuhiro
IEICE technical report   106(300) 37-44   Oct 2006
To realize natural human-robot interaction, we consider that a robot should have at least two functions, that is, real-world auditory scene analysis by a robot to understand the surrounding environments, and robot expression to send information to...
SUMIYA Ryota, NAKADAI Kazuhiro, NAKANO Mikio, ICHIGE Koichi, HIROSE Yasuo, TSUJINO Hiroshi
Proceedings of the IEICE General Conference   2006(1)    Mar 2006
KANDA NAOYUKI, KOMATANI KAZUNORI, NAKANO MIKIO, NAKADAI KAZUHIRO, TSUJINO HIROSHI, OGATA TETSUYA, OKUNO HIROSHI G.
IPSJ SIG Notes   2006(12) 55-60   Feb 2006
We have developed a robust domain selection method using dialogue history in multi-domain spoken dialogue systems. We define domain selection as classifying problem among (I) the domain in the previous turn, (II) the domain in which N-best speech ...
TSUJINO Hiroshi, NAKANO Mikio, NAKADAI Kazuhiro, HASEGAWA Yuji
IEICE technical report   105(426) 31-36   Nov 2005
As the computer technology advances, machines are expected to perform more functional tasks at home and the importance of technology realizing "human-machine interface that anyone can use" is increasing. An intelligent robot is an ultimate machine...
KUROTAKI Shunsuke, SUZUKI Noriaki, NAKADAI Kazuhiro, OKUNO Hiroshi, AMANO Hideharu
IEICE technical report   105(43) 67-72   May 2005
OKUNO Hiroshi G., NAKADAI Kazuhiro
日本音響学会研究発表会講演論文集   2005(1) 633-636   Mar 2005
Yamamoto Shunichi, Nakadai Kazuhiro, Tsujino Hiroshi, Okuno Hiroshi G.
情報科学技術フォーラム一般講演論文集   3(2) 357-360   Aug 2004
SUZUKI Noriaki, NAKADAI Kazuhiro, AMANO Hideharu, OKUNO Hiroshi G., KITANO Hiroaki
情報処理学会研究報告システムLSI設計技術(SLDM)   2003(7) 135-140   Jan 2003
Reconfigurable systems are efficient for high performance but low cost/power implementation for intelligent systems for robots. In this paper, a part of processing for the direction-pass filter, such as Fast Fourier Transform(FFT), square root, an...
SUZUKI Noriaki, NAKADAI Kazuhiro, AMANO Hideharu, OKUNO Hiroshi G., KITANO Hiroaki
IEICE technical report. Computer systems   102(611) 79-84   Jan 2003
Reconfigurable systems are efficient for high performance but low cost/power implementation for intelligent systems for robots. In this paper, a part of processing for the direction-pass filter, such as Fast Fourier Transform (FFT), square root, a...
SUZUKI Noriaki, NAKADAI Kazuhiro, AMANO Hideharu, OKUNO Hiroshi G., KITANO Hiroaki
Technical report of IEICE. VLD   102(609) 79-84   Jan 2003
Reconfigurable systems are efficient for high performance but low cost/power implementation for intelligent systems for robots. In this paper, a part of processing for the direction-pass filter, such as Fast Fourier Transform (FFT), square root, a...
OKUNO Hiroshi G., NAKADAI Kazuhiro
IPSJ SIG Notes   2001(123) 69-74   Dec 2001
In this paper, we present an active audition system which is implemented on the humanoid robot "SIG the humanoid". The audition system for highly intelligent humanoids localize sound sources and recognize auditory events in the auditory scene. Act...
OKUNO Hiroshi G., NAKADAI Kazuhiro
IEICE technical report. Natural language understanding and models of communication   101(520) 69-74   Dec 2001
In this paper, we present an active audition system which is implemented on the humanoid robot "SIG the humanoid". The audition system for highly intelligent humanoids localize sound sources and recognize auditory events in the auditory scene. Act...
OKUNO Hiroshi G., NAKADAI Kazuhiro
IEICE technical report. Speech   101(522) 69-74   Dec 2001
In this paper, we present an active audition system which is implemented on the humanoid robot "SIG the humanoid". The audition system for highly intelligent humanoids localize sound sources and recognize auditory events in the auditory scene. Act...
NAKADAI KAZUHIRO, HIDAI KEN-ICHI, OKUNO HIROSHI G., KITANO HIROAKI
IPSJ SIG Notes. ICS   2001(97) 37-42   Oct 2001
This paper describes improvement of auditory processing by active motion and audio-visual integration. Generally, environmental noises and reverberation affect sound source localization and separation in the real world badly. Our real-time human t...
OKUNO HIROSHI G., KYODA KOJI M., NAKADAI KAZUHIRO, KITANO HIROAKI
IPSJ SIG Notes   2000(23) 119-124   Mar 2000
Beowulf-Class cluster is a logical organization of PC clusters composed of mass-market off-the-shelf hardware and software. The user may have problems that their implementation won't work well in hardware level or their implementation provides qui...
OKUNO HIROSHI G., KYODA KOJI M., NAKADAI KAZUHIRO, KITANO HIROAKI
IPSJ SIG Notes   2000(23) 116-124   Mar 2000
Beowulf-Class cluster is a logical organization of PC clusters composed of mass-market off-the-shelf hardware and software. The user may have problems that their implementation won't work well in hardware level or their implementation provides qui...
HANDA Ibuki, KINOSHITA Tomoyoshi, MUTO Makoto, SAKAI Shuichi, TANAKA Hidehiko
IPSJ SIG Notes   2000(19) 21-26   Feb 2000
We have proposed OPTIMA, a system for music scene analysis. The system sometimes misses musical notes. It can be said that complete transcription is difficult for a music transcription system which depends on only computational processes. So, we p...
HANDA Ibuki, KINOSHITA Tomoyoshi, MUTO Makoto, SAKAI Shuichi, TANAKA Hidehiko
IEICE technical report. Speech   99(626) 21-26   Feb 2000
We have proposed OPTIMA, a system for music scene analysis. The system sometimes misses musical notes. It can be said that complete transcription is difficult for a music transcription system wchich depends on only computational processes. So, we ...
武藤 誠, 木下 智義, 半田 伊吹, 坂井 修一, 田中 英彦
全国大会講演論文集   59(0) 11-12   Sep 1999
半田 伊吹, 木下 智義, 武藤 誠, 坂井 修一, 田中 英彦
全国大会講演論文集   59(0) 25-26   Sep 1999
木下 智義, 半田 伊吹, 武藤 誠, 坂井 修一, 田中 英彦
全国大会講演論文集   59(0) 27-28   Sep 1999
KINOSHITA Tomoyoshi, SAKAI Shuichi, TANAKA Hidehiko
IEICE technical report. Speech   98(611) 1-6   Feb 1999
We have previously proposed a processing model OPTIMA for music scene analysis and implemented its experimental system. However, the system was not robust to signals with overlapped frequency components. In this paper, we present a new method that...
KINOSHITA Tomoyoshi, SAKAI Shuichi, TANAKA Hidehiko
IPSJ SIG Notes   1999(16) 49-54   Feb 1999
We have previously proposed a processing model OPTIMA for music scene analysis and implemented its experimental system. However, the system was not robust to signals with overlapped frequency components. In this paper, we present a new method that...
WATANABE Hiroshi, NAKADAI Kazuhiro, SATOU Yukio, SAKAGUCHI Zenji, ASHIKAWA Hirotoshi
IEICE technical report. Computer systems   98(572) 1-8   Jan 1999
For operating the reliable data communications, we use the protocol message to control that communications. In case of the continuous and high-speed operating message to the node on purpose, it arises the problem that we cannot offer the service f...
木下 智義, 村岡 秀哉, 田中 英彦
全国大会講演論文集   56(0) 32-33   Mar 1998
MURAOKA Hideya, KINOSHITA Tomoyoshi, TANAKA Hidehiko
Proceedings of the IEICE General Conference   1998(1)    Mar 1998
KINOSHITA Tomoyoshi, TANAKA Hidehiko, Tomoyoshi Kinoshita, Hidehiko Tanaka, University of Tokyo Department of Electorical Engineering, University of Tokyo Department of Electorical Engineering
コンピュータソフトウェア   15(1) 59-66   Jan 1998
Kinoshita Tomoyoshi, Muraoka Hideya, Tanaka Hidehiko
THE JOURNAL OF THE ACOUSTICAL SOCIETY OF JAPAN   54(3) 190-198   1998
筆者らは, 知覚的な音の階層的な内部像を構築するために, これまでに音楽情景分析の処理モデルOPTIMAを提案し, その実験システムを実装した。しかしながら, そのシステムは認識率の面で実用的とは言えず, その改善が課題となっている。OPTIMAは, 複数の処理モジュールからの出力を統合して外界の最尤推定像を得る枠組であるため, 処理モジュールの追加によって認識精度の向上を図ることができる。本稿ではまず単音の遷移に注目し, 単音の遷移に関する統計的分析を行って知識源を構成した。次に, 単音...
村岡 秀哉, 木下 智義, 田中 英彦
全国大会講演論文集   55(0) 29-30   Sep 1997
複数の音が混在する音響信号から, 外界の状況を認識理解するという人間の聴覚における情報処理を聴覚的情景分析と呼び, その計算機上での実現の研究がさかんに行われている。従来の研究の多くは, 音声あるいは音楽といった特定の入力の混合音を前提として分離・認識するものであり, 音一般に関する研究はあまり見られない。これは, 高精度な認識処理を行うために特定の音にあてはまるルール/知識を処理システムに導入している点が一因にあると考えられる。さて混合音の認識においては, 入力信号から個々の音に相当する...
KINOSHITA Tomoyoshi, MURAOKA Hideya, TANAKA Hidehiko
IEICE technical report. Speech   96(540) 15-20   Feb 1997
Previously, we have proposed a processing model OPTIMA for music scene analysis, and implemented its experimental system. However, its recognition accuracy was not practical. The OPTIMA processing architecture is the framework where multiple sourc...
KINOSHITA Tomoyoshi, MURAOKA Hideya, TANAKA Hidehiko
IPSJ SIG Notes   1997(18) 49-54   Feb 1997
Previously, we have proposed a processing model OPTIMA for music scene analysis, and implemented its experimental system. However, its recognition accuracy was not practical. The OPTIMA processing architecture is the framework where multiple sourc...
木下 智義, 村岡 秀哉, 田中 英彦
全国大会講演論文集   53(0) 369-370   Sep 1996
著者らは音楽情景分析の処理モデルOPTIMAを提案し、その実験システムを実装した。OPTIMAでは、複数の独立した処理モジュールを用意し、確率をもった仮説の組と、これらの仮説の組みの間の条件つき確率を出力させ、これらを用いて仮説ネットワークを構成する。仮説の組みはノードとして、条件つき確率はリンクとして表される。その後、確率伝搬によって確率情報を統合することにより、外界の音響的事象に関する最尤推定像を求める。現時点では実装したOPTIMA実験システムの認識率は実用に十分であるとは言えない。...
村岡 秀哉, 木下 智義, 田中 英彦
全国大会講演論文集   52(0) 457-458   Mar 1996
著者らは音楽情景分析の処理モデルOPTIMAを実装している。OPTIMAでは、複数の独立したモジュールに確率を持った仮説の組を出力させ、これを確率伝搬によって統合することにより、外界の音響的事象に関する最尤推定像を求める。
木下 智義, 田中 英彦
全国大会講演論文集   51(0) 279-280   Sep 1995
著者らは音楽情景分析の処理モデルOPTIMAを実装した[?,?,?]。OPTIMAでは、複数の独立したモジュールに確率をもった仮説の組を出力させ、これを確率伝搬によって統合することによって外界の音響的事象に関する最尤推定像を求める。現時点では実装したOPTIMA実験システムの認識率は実用に十分であるとは言えず、その改善が課題となっている。本稿ではOPTIMAにおける誤認識の原因を解析し、認識率改善のための手法について考察する。
柏野 邦夫, 中臺 一博, 木下 智義, 田中 英彦
全国大会講演論文集   50(0) 97-98   Mar 1995
われわれは、聴覚的情景分析を「知覚的な音」の分離抽出(知覚的音源分離)と構造化の問題と捉え、モノラルの楽器演奏の音響信号を題材として、音楽情景分析(音楽音響信号を対象とする聴覚的情景分析)の処理モデルについて検討を行っている。ここで、知覚的音源分離とは、人間がひとつのものとして知覚または認識するような音響エネルギーのまとまり(これを知覚的な音と呼ぶ)を一つのものとして記号化することを指す。われわれは既に、ベイズの定理に基礎を置く定量的かつ階層的な情報統合のメカニズムを備えた音楽情景分析の処...
木下 智義, 柏野 邦夫, 中臺 一博, 田中 英彦
全国大会講演論文集   50(0) 99-100   Mar 1995
OPTIMAでは、複数の独立したモジュールに確率をもった仮説の組を出力させ、これを確率伝搬によって統合することによって外界の音響的事象に関する最尤推定像を求める。本稿ではOPTIMAにおいて利用される音楽シーン惰報として、拍位置および和音の情報の抽出と利用について議論し、実験システムに対する評価実験の結果を示す。
中臺 一博, 柏野 邦夫, 木下 智義, 田中 英彦
全国大会講演論文集   50(0) 101-102   Mar 1995
われわれは、音楽情景分析における処理モデルとしてOPTIMAを提案し、これに基づく音楽情景分析の実験システムの実装・評価を行った。本稿では、実験システムのうち、周波数成分レベル、単音レベル間の処理を行う単音仮説生成処理部の実装および、評価について述べる。
NAKADAI Kazuhiro, KASHINO Kunio, KINOSHITA Tomoyoshi, TANAKA Hidehiko
日本音響学会研究発表会講演論文集   1995(1) 481-482   Mar 1995
KASHINO Kunio, NAKADAI Kazuhiro, KINOSHITA Tomoyoshi, TANAKA Hidehiko
日本音響学会研究発表会講演論文集   1995(1) 483-484   Mar 1995
柏野 邦夫, 中台 一博, 田中 英彦
全国大会講演論文集   49(0) 325-326   Sep 1994
われわれは、モノラルの楽器演奏を対象とする音源分離を題材として、知覚的音源分離システムについて検討を進めている。知覚的音源分離においては、観測データに加え、対象に関する知識や記憶に基づく処理を柔軟に組み合わせて最終的な結果を求めることが本質的な課題である。そこで本稿では、情報統合のメカニズムを備えた知覚的音源分離の処理モデル OPTIMA (Organized Processing toward Intelligent Music Scene Analysis)を提案する。

Conference Activities & Talks

 
Nelson Yalta, Shinji Watanabe, Takaaki Hori, Kazuhiro Nakadai, Tetsuya Ogata
7 Nov 2018   
Casual conversations involving multiple speakers and noises from surrounding
devices are part of everyday environments and pose challenges for automatic
speech recognition systems. These challenges in speech recognition are target
for the CHiME-5 ...
尾崎翔, 浅野太, 中臺一博
電子情報通信学会大会講演論文集(CD-ROM)   28 Aug 2018   
糸山克寿, 中臺一博, 中臺一博
日本機械学会ロボティクス・メカトロニクス講演会講演論文集(CD-ROM)   1 Jun 2018   
奥乃博, 糸山克寿, 中臺一博, 中臺一博, 公文誠, 坂東宜昭, 干場功太郎
システム制御情報学会研究発表講演会講演論文集(CD-ROM)   16 May 2018   
谷口亮輔, 干場功太郎, 中臺一博, 中臺一博
情報処理学会全国大会講演論文集   13 Mar 2018