Updated on 2023/11/30

写真a

 
OTSUKI Takashi
 
Title
Associate Professor

Research Areas

  • Informatics / Perceptual information processing

  • Informatics / Perceptual information processing

 

Papers

  • L1 Vocabulary 3D Map (L1 Map): A Visual Map of L1 Word Relations in English through Word Embedding, Proceedings of the Annual Conference of JSAI, JSAI2023(0) 1T3GS603, 2023

    KANEKO Jun, OTSUKI Takashi, SAKAGUCHI Takayuki, SOKOLOVSKY Jesse

    Multiple Authorship (Only Japanese)

  • AIWolf Agent Using DNN to Predict Voting Behavior of Others, Proceedings of the Annual Conference of JSAI, JSAI2023(0) 4Xin165, 2023

    OTSUKI Takashi, EBE Hirotaka

    Multiple Authorship (Only Japanese)

  • Diachronic Connotation Analysis of English and American Literature Based on the Similarity Between Word Vectors, Proceedings of the Annual Conference of JSAI, JSAI2022(0) 1K1GS605, 2022

    KANEKO Jun, OTSUKI Takashi, SAKAGUCHI Takayuki

    Multiple Authorship (Only Japanese)

  • Negotiation in Hidden Identity: Designing Protocol for Werewolf Game, Recent Advances in Agent-based Negotiation. ACAN 2019. Studies in Computational Intelligence, vol 958, Springer, Singapore, 87-102, 2021.05

    Hirotaka Osawa, Takashi Otsuki, Claus Aranha, Fujio Toriumi

    Multiple Authorship (Only Japanese)

  • An Approach to the Study of English and American Literature Using Text Mining and Word Vectors:Interpreting Herman Melville's <i>Typee</i> from the Perspective of Natural Language Processing, Proceedings of the Annual Conference of JSAI, 2021(0) 4J4GS6g02, 2021

    KANEKO Jun, OTSUKI Takashi, SAKAGUCHI Takayuki

    Multiple Authorship (Only Japanese)

  • AI Analysis of Junior and Senior High School Teachers’ Awareness of the Performance Test, Implementing the Latest Natural Language Processing Technology with Python, 東北英語教育学会研究紀要, (41) 129-144, 2021

    Jun KANEKO, Takashi OTSUKI, Takayuki SAKAGUCHI, Jerry MILLER, Tsuneo YAMAGUCHI, Eiji KOBAYASHI

    Multiple Authorship (Only Japanese)

  • Effect of meta-inference using utterance patterns in role estimation of AIWolf, JSAI2021 1J2GS10d01, 2021

    OTSUKI Takashi

    Single Author

  • Evaluation of the AI-based werewolf game player agent which can persuade and can be persuaded, JSAI2020 1P3GS705, 2020

    OTSUKI Takashi, KONTA Yukiya

    Multiple Authorship (Only Japanese)

  • Overview of AIWolfDial 2019 Shared Task: Contest of Automatic Dialog Agents to Play the Werewolf Game through Conversations, Proceedings of the 1st International Workshop of AI Werewolf and Dialog System (AIWolfDial2019), 1-6, 2019.10

    Yoshinobu Kano , Claus Aranha, Michimasa Inaba, Fujio Toriumi, Hirotaka Osawa, Daisuke Katagami, Takashi Otsuki, Issei Tsunoda, Shoji Nagayama, Dolça Tellols, Yu Sugawara, Yohei Nakata

    Multiple Authorship (Only Japanese)

  • Evaluation of topic n-gram model for the utterance of AIWolf agent, Proceedings of the Annual Conference of JSAI, 2018(0) 1H2OS13b01, 2018

    OTSUKI Takashi

    Single Author

  • Annotating Expression of Roles and Exercise of Special Abilities to the Werewolf BBS, Proceedings of the Annual Conference of JSAI, 2018(0) 1H1OS13a01, 2018

    INABA Michimasa, KANO Yoshinobu, OSAWA Hirotaka, OTSUKI Takeshi, KATAGAMI Daisuke, TORIUMI Fujio

    Multiple Authorship (Only Japanese)

  • Improvement of team estimation in AIWolf player using machine learning , 2017 64-69, 2017

    Multiple Authorship (Only Japanese)

  • Prediction Method for Word Intelligibility Based on Word Familiarity, Proceedings of the Auditory Research Meeting, 40(6) 483-488, 2010.07

    Multiple Authorship (Only Japanese)

  • The optimization of reference pixel arrangement in the lossless coding for HDTV image, IEICE Technical Report, 109(469) 91-96, 2010.03

    Multiple Authorship (Only Japanese)

  • An effecitive optimization of reference pixel arrangement in the lossless coding for HDTV image, 2010 IEEE/ACIS 9th International Conference on Computer and Information Science, 246-251, 2010

    Yuya Yamaguchi, Takashi Otsuki, Ken-ichi Iwata, Shuichit Itoh

    Multiple Authorship (Only Japanese)

  • A lossless coding for still image of HDTV using adaptive liner prediction, IEICE Technical Report, 109(226) 35-40, 2009.10

    Multiple Authorship (Only Japanese)

  • Prediction Formula of Optimal Language Weight on Sentence Recognition, 8(2) 355-356, 2009.08

    Single Author

  • Performance Prediction on Sentence Recognition Using Statistical Language Models, Proceedings of Oriental-COCOSDA 2008, 205-208, 2008.11

    Takashi OTSUKI, Shozo MAKINO

    Multiple Authorship (Only Japanese)

  • Cell-Free Boundary Node Method, 94-95, 2007.09

    Multiple Authorship (Only Japanese)

  • Automatic Lipreading of Station Names Using Optical Flow and HMM, IEICE Technical Report, 102(473) 25-30, 2002.11

    Multiple Authorship (Only Japanese)

  • Self-Generation of Gait Pattern in Quadruped Walking Robot Using Vector Quantizing Search Method (VQSM), Transactions of the Institute of Systems, Control and Information Engineers, 15(8) 396-403, 2002.08

    Multiple Authorship (Only Japanese)

  • Construction of Segmental Unit Input with Dual-Width Window for Speech Recogniton, Transactions of the Institute of Systems, Control and Information Engineers, 14(11) 522-529, 2001.11

    Multiple Authorship (Only Japanese)

  • Travel Control for Robot Vehicle Using Neural Networks in Dynamic Environment, Transactions of the Institute of Systems, Control and Information Engineers, 14(2) 52-61, 2001.02

    Multiple Authorship (Only Japanese)

  • Language Model Evaluation Method Based on Hamming Distance between Sentences, IEICE Technical Report, 100(137) 61-66, 2000.06

    Multiple Authorship (Only Japanese)

  • Calculation Time Reduction for Path Planning of Robot Vehicle Using Neural Networks, Transactions of the Institute of Systems, Control and Information Engineers, 13(5) 250-257, 2000.05

    Multiple Authorship (Only Japanese)

  • The investigation of the language model evaluation method based on the Hamming distance between sentences, Reports of the Meeting. the Acoustical Society of Japan, 1999(Autumn Pt 1) 75-76, 1999.09

    Multiple Authorship (Only Japanese)

  • Dual-Width Windowed Segment for Speech Recognition, Proceedings of 1999 Ingternational Conference on Speech Processing(ICSP '99), 91-95, 1999.08

    Takashi Otsuki, Kazunori Ninomiya, Teruhiko Ohtomo

    Multiple Authorship (Only Japanese)

  • The Examination of Variable-Length Segment Unit Input in the Speech Recognition, Proceedings of 1999 Ingternational Conference on Speech Processing(ICSP '99), 283-288, 1999.08

    Kazunori NINOMIYA, Takashi OTSUKI, Teruhiko OHTOMO

    Multiple Authorship (Only Japanese)

  • Face Discrimination Using Neural Network under Varying View, Transactions of the Institute of Systems, Control and Information Engineers, 12(3) 161, 1999.03

    Multiple Authorship (Only Japanese)

  • A Study of Compound Segmental Unit Input in Discrete HMM, Reports of the Meeting. the Acoustical Society of Japan, 1999(Spring Pt 1) 11-12, 1999.03

    Multiple Authorship (Only Japanese)

  • The 3rd Report on Investigation of Feature Vector for Speech Recognition, Reports of the Meeting. the Acoustical Society of Japan, 1998(Autumn Pt 1) 13-14, 1998.09

    Multiple Authorship (Only Japanese)

  • On the Shape of Segmental Unit Input for Speech Recognition, Reports of the Meeting. the Acoustical Society of Japan, 1998(Autumn Pt 1) 11-12, 1998.09

    Multiple Authorship (Only Japanese)

  • The Property of Asymmetric Segment, IEICE Technical Report, 98(106) 53-59, 1998.06

    Multiple Authorship (Only Japanese)

  • The Efficiency of the Dual-Width Windowed Segment on Speech Recognition, IEICE Technical Report, 97(561) 63-70, 1998.02

    Multiple Authorship (Only Japanese)

  • On Feature Vector for Speech Recognition, IEICE Technical Report, 97(396) 27-34, 1997.11

    Multiple Authorship (Only Japanese)

  • On Effects of DWWS for Discrete HMM Speech Recognition System, Reports of the Meeting. the Acoustical Society of Japan, 1997(Autumn Pt 1) 67-68, 1997.09

    Multiple Authorship (Only Japanese)

  • The 2nd Report on Investigation of Feature Vector for Speech Recognition, Reports of the Meeting. the Acoustical Society of Japan, 1997(Autumn Pt 1) 11-12, 1997.09

    Multiple Authorship (Only Japanese)

  • The 1st Report on Investigation of Feature Vector for Speech Recognition, Reports of the Meeting. the Acoustical Society of Japan, 1996(Autumn Pt 1) 61-62, 1996.09

    Multiple Authorship (Only Japanese)

  • Study of DWWS-LVQ Speech Recognition System with Multi-Tempulate, Reports of the Meeting. the Acoustical Society of Japan, 1996(Autumn Pt 1) 79-80, 1996.09

    Multiple Authorship (Only Japanese)

  • Verification of the effectiveness of a two-step window segment unit by the degree of separation between phonemes classes, Tohoku-Section Joint Convention Record on Institutes of Electrical and Information Engineers, Japan, 1996 144, 1996.08

    Multiple Authorship (Only Japanese)

  • Study of LVQ Speech Recognition System Using Segmental Unit Input with Dual-Width Window, IEICE Technical Report, 96(93) 53-60, 1996.06

    Multiple Authorship (Only Japanese)

  • The Property of Phoneme Labels Given by Auto-Study, IEICE Technical Report, 96(93) 39-46, 1996.06

    Multiple Authorship (Only Japanese)

  • The Performance Prediction on Sentence Recognition Using a Finite State Word Automaton, IEICE transactions on information and systems, 79(1) 47-53, 1996.01

    Takashi OTSUKI, Akinori ITO, Shozo MAKINO, Teruhiko OHTOMO

    Multiple Authorship (Only Japanese)

  • Improvement of BP Algorithm for Neural Network and Its Evaluation due to Recognition of Hand-written Chinese Characters, Proceedings of ICONIP '95, 1995.11

    Teruhiko Ohtomo, Qing Yang, Takashi Otsuki

    Multiple Authorship (Only Japanese)

  • Obstacle Avoidance Travel Control of Robot Vehicle Using Multiple Neural Networks, Proceedings of ICONIP '95, 1995.11

    Minoru Kodaira, Teruhiko Ohtomo, Takashi Otsuki, Takao Ohuchi

    Multiple Authorship (Only Japanese)

  • Obstacle Avoidance Travel Control of Robot Vehicle Using Multiple Cascaded Neural Networks, 173-174, 1995.10

    Multiple Authorship (Only Japanese)

  • On Effects of Auto Study for LVQ Speech Recognition System, Reports of the Meeting. the Acoustical Society of Japan, 1995(Autumn Pt 1) 159-160, 1995.09

    Multiple Authorship (Only Japanese)

  • Effect of .DELTA. learning process in phoneme recognition using recursive network, Tohoku-Section Joint Convention Record on Institutes of Electrical and Information Engineers, Japan, 1995 9, 1995.08

    Multiple Authorship (Only Japanese)

  • Improvement of Learning algorithm for Locally Connected Neural Networks and Three-Stage Recognition Method of Hand-Written Characters, IEICE Technical Report, 95(43) 25-32, 1995.05

    Multiple Authorship (Only Japanese)

  • On-line Writer Discrimination of Hand-Written Characters Using P-type Fourier Discriptor, IPSJ Journal, 35(11) 2300-2307, 1994.11

    Multiple Authorship (Only Japanese)

  • Performance prediction of word recognition using the transition information between phonemes or between characters, Systems and Computers in Japan, 25(7) 72-81, 1994.09

    Takashi Otsuki, Shozo Makino, Akinori Ito, Toshio Sone

    Multiple Authorship (Only Japanese)

  • On Efficient Recognition of Hand-written Chinese Characters by Locally Connected Neural Networks, IPSJ Journal, 35(6) 1091-1100, 1994.06

    Multiple Authorship (Only Japanese)

  • Performance Prediction of Word Recognition Using the Probability of Word Occurrence, IEICE Transaction on Fundamentals of Electronics, Communications and Computer Sciences (Japan Edition), 77(2) 274-281, 1994.02

    Multiple Authorship (Only Japanese)

  • The performance prediction method on sentence recognition system using a finite state automaton, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing, 1 397-400, 1994

    Takashi Otsuki, Akinori Ito, Shozo Makino, Teruhiko Ohtomo

    Multiple Authorship (Only Japanese)

  • Performance Estimation of Word Recognition System Using the Probability of Word Occurrence, 1993(Shuki Pt 1) 142, 1993.09

    Multiple Authorship (Only Japanese)

  • Writer verification of hand-written characters using P-type fourier descriptor, 1993.09

    Multiple Authorship (Only Japanese)

  • Off-Line Recognition of Chinese Characters Using P-Type Fourier Descriptor, 1993(Shuki Pt 6) 306, 1993.09

    Multiple Authorship (Only Japanese)

  • Recognition of Vowel Sequences Using a Neural Network Model, 1993(Shuki Pt 6) 290, 1993.09

    Multiple Authorship (Only Japanese)

  • The Performance Evaluation on Sentence Recognition System Using A Finite State Automaton - The Relationship between Word Recognition Score and Sentence Recognition Score, IEICE Technical Report, 93(183) 41-48, 1993.08

    Multiple Authorship (Only Japanese)

  • Performance Prediction of Word Recognition Using the Transition Information between Phonemes or between Characters, IEICE Transactions on Information and Systems, Pt.2 (Japanese Edition), 76(6) 1090-1096, 1993.06

    Multiple Authorship (Only Japanese)

  • The performance evaluation method on sentence recognition system using a finite state automaton, Reports of the Meeting. the Acoustical Society of Japan, 1993(Spring Pt 1) 73-74, 1993.03

    Multiple Authorship (Only Japanese)

  • The performance evaluation method on sentence recognition system which uses the transition information between word categories, 46th(2) 197-198, 1993.03

    Multiple Authorship (Only Japanese)

  • An accurate formula to calculate word recognition score, Reports of the Meeting. the Acoustical Society of Japan, 1992(Autumn Pt 1) 183-184, 1992.10

    Multiple Authorship (Only Japanese)

  • The relationship between the phoneme recognition score and the word recognition score in the various tasks, Reports of the Meeting. the Acoustical Society of Japan, 1992(Spring Pt 1) 35-36, 1992.03

    Multiple Authorship (Only Japanese)

  • The relationship between the character recognition score and the word recognition score on character recognition using linguistic information, 44th(2) 141-142, 1992.02

    Multiple Authorship (Only Japanese)

  • The relationship between the phoneme recognition score and the word recognition score on speech recognition using a priori probability of word, Reports of the Meeting. the Acoustical Society of Japan, 1991(Autumn Pt 1) 169-170, 1991.10

    Multiple Authorship (Only Japanese)

  • Performance evaluation in word recognition using transition information between phonemes - comparison between theoretical and simulation-derived recognition scores -, Proceeding of Korea-Japan joint workshop on advanced technology of speech recognition and synthesis, 68-73, 1991.07

    Takashi Otsuki, Shozo Makino, Toshio Sone, Ken'ichi Kido

    Multiple Authorship (Only Japanese)

  • Performance evaluation in speech recognition system using transition probability between linguistic units, Proceedings of ICSLP '90, 1213-1216, 1990.11

    Takashi Otsuki, Shozo Makino, Toshio Sone, Ken'ichi Kido

    Multiple Authorship (Only Japanese)

  • Comparison between theoretical and simulation-derived recognition scores in word recognition using transition information between phonemes, Reports of the Meeting. the Acoustical Society of Japan, 1990(Autumn Pt 1) 115-116, 1990.09

    Multiple Authorship (Only Japanese)

  • Performance evaluation of speech recognition method using transition information between linguistic units, IEICE Technical Report, 90(112) 33-38, 1990.06

    Multiple Authorship (Only Japanese)

  • Performance evaluation of word recognition method using transition information between phonemes, Reports of the Meeting. the Acoustical Society of Japan, 1990(Spring Pt 1) 73-74, 1990.03

    Multiple Authorship (Only Japanese)

  • Performance prediction of character and speech recognition using linguistic information, Reports of the Meeting. the Acoustical Society of Japan, 1989(Autumn Pt 1) 111-112, 1989.10

    Multiple Authorship (Only Japanese)

  • An investigation on efficiency of trigram on Japanese spoken sentence recognition, Reports of the Meeting. the Acoustical Society of Japan, 1989(Spring Pt 1) 99-100, 1989.03

    Multiple Authorship (Only Japanese)

display all >>

Books

  • Programming AI Werewolf Game Player , Mynavi Publishing Corporation, 2017.06

    KANO Yoshinobu,OTSUKI Takashi,SONODA Atom,NAKATA Yohei,MINOWA Takashi,TORIUMI Fujio

  • Computer Processing of Asian Spoken Languages, Consideration Books, 2010.03

    Shuichi Itahashi 他

  • Insect Mimetics, 2008.10

Review Papers

  • Summary of the OS : AI Werewolf and Incomplete Information Games, The Japanese Society for Artificial Intelligence, Journal of the Japanese Society for Artificial Intelligence, 36(5) 613-617, 2021

  • AI Wolf Contest : After the International Contest and the Natural Language Contest, The Japanese Society for Artificial Intelligence, Journal of the Japanese Society for Artificial Intelligence, 35(3) 344-350, 2020

  • AI Werewolf and Incomplete Information Games, The Japanese Society for Artificial Intelligence, Journal of the Japanese Society for Artificial Intelligence, (34) 876-880, 2019.11

  • Introduction to Research on Artificial Intelligence Based Werewolf, Japan Society for Fuzzy Theory and Intelligent Informatics, Journal of Japan Society for Fuzzy Theory and Intelligent Informatics, 30(5) 236-244, 2018.10

    KATAGAMI Daisuke, TORIUMI Fujio, OSAWA Hirotaka, KANO Yoshinobu, INABA Michimasa, OTSUKI Takashi

Grant-in-Aid for Scientific Research

  • Grant-in-Aid for Scientific Research(C),2023.04 - 2024.03

  • Grant-in-Aid for Scientific Research(C),2020.04 - 2023.03

Presentations

  • Annual Conference of JSAI, Domestic presentation, 2020.06, Evaluation of the AI-based werewolf game player agent which can persuade and can be persuaded, Oral presentation (general)

  • ENTERTAINMENT COMPUTING 2017, Domestic presentation, 2017.09, Oral presentation (general)

  • The 1993 IEICE Fall Conference, Domestic presentation, 1993.09, Performance Estimation of Word Recognition System Using the Probability of Word Occurrence, Oral presentation (general)

  • The 1993 IEICE Fall Conference, Domestic presentation, 1993.09, Recognition of Vowel Sequences Using a Neural Network Model, Oral presentation (general)

  • The 1993 IEICE Fall Conference, Domestic presentation, 1993.09, Off-Line Recognition of Chinese Characters Using P-Type Fourier Descriptor, Oral presentation (general)

  • The 1993 IEICE Fall Conference, Domestic presentation, 1993.09, Writer verification of hand-written characters using P-type fourier descriptor, Oral presentation (general)

display all >>