Updated on 2024/03/05

写真a

 
KOSAKA Tetsuo
 
Title
Professor

Research Areas

  • Informatics / Perceptual information processing

Graduating School

  • Tohoku University, Faculty of Engineering

    1984.03, Graduated

Graduate School

  • Tohoku University, Graduate School, Division of Engineering

    Master's Course, 1986.03, Completed

 

Papers

  • Simultaneous Adaptation of Acoustic and Language Models for Emotional Speech Recognition Using Tweet Data, IEICE Trans. Information ans Systems, E107-D(3) 363-373, 2024.03

    Tetsuo Kosaka, Kazuya Saeki, Yoshitaka Aizawa, Masaharu Kato, Takashi Nose

    Multiple Authorship (Only Japanese)

  • Speech emotion recognition by late fusion of linguistic and acoustic features using deep learning models, Proc. of APSIPA ASC 2023, 1013-1018, 2023.11

    Kiyohide Sato, Keita Kishi, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Frequency resolution improvement by sub-bins and an inverse matrix in discrete Fourier transform, Applied optics, 62(25) 6614-6624, 2023.08

    Manabu Sato, Shunta Nakabayashi, Ryusei Ito, Junpei Masuta, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Emotion recognition combininb acoustic and linguistic features based on speech recognition results, Proc. of 2021 IEEE 10th Global Conference on Consumer Electronics, 889-892, 2021.10

    Misaki Sakurai, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Improvement of frequency resolution using a sub-bin structure in the discrete Fourier transform, Applied optics, 60(21) 6290-6301, 2021.07

    Manabu Sato, Yuuki Kimura, Junpei Masuta, Tetsuo Kosaka, Izumi Nishidate

    Multiple Authorship (Only Japanese)

  • Language Model Adaptation for Emotional Speech Recognition using Tweet Data, Proc. of APSIPA ASC 2020, 371-375, 2020.12

    Kazuya Saeki, Masaharu Kato, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Performance Improvement of Prosody-Controlled Voice Conversion by Language Model Adaptation, Proc. of 2019 IEEE 8th Global Conference on Consumer Electronics, 854-856, 2019.10

    Kazuya Saeki, Masaharu Kato, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Investigation of Acoustic Models for Emotion Recognition Using a Spontaneous Speech Corpus, Proc. of 23nd International Congress on Acoustics, 0300, 2019.09

    Tetsuo Kosaka, Yuka Haneda, Daisuke Makabe, Masaharu Kato,

    Multiple Authorship (Only Japanese)

  • Acoustic Model Adaptation for Emotional Speech Recognition Using Twitter-Based Emotional Speech Corpus, Proc. of APSIPA ASC 2018, 1747-1751, 2018.11

    Tetsuo Kosaka, Yoshitaka Aizawa, Masaharu Kato, Takashi Nose

    Multiple Authorship (Only Japanese)

  • Improving Voice Activity Detection for Multimodal Movie Dialogue Corpus, Proc. of 2018 IEEE 7th Global Conference on Consumer Electronics, 453-456, 2018.10

    Tetsuo Kosaka, Ikumi Suga, Masashi Inoue

    Multiple Authorship (Only Japanese)

  • Multimodal Spoken Dialog System Using State Estimation by Body Motion, Proc. of 2017 IEEE 6th Global Conference on Consumer Electronics, 348-351, 2017.10

    Takeru Koseki, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Voice activity detection in movies using multi-class deep neural networks, 5th Joint Meeting of the Acoustical Society of America and Acoustical Society of Japan, 2016.12

    I. Suga, R. Yasuhara, M. Inoue and T. Kosaka

    Multiple Authorship (Only Japanese)

  • Many-to-many voice conversion using hidden Markov model-based speech recognition and synthesis, 5th Joint Meeting of the Acoustical Society of America and Acoustical Society of Japan, 2016.12

    Y. Aizawa, M. Kato and T. Kosaka

    Multiple Authorship (Only Japanese)

  • Large-scale multimodal movie dialogue corpus, Proc. of the 18th ACM International Conference on Multimodal Interaction, 414-415, 2016.11

    Ryu Yasuhara, Masashi Inoue, Ikumi Suga and Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Voice Conversion of emotional speech using hidden Markov model-based speech recognition and synthesis, Proc. of 22nd International Congress on Acoustics, 0300, 2016.09

    Tetsuo Kosaka, Yoshiaki Nakagawa and Masaharu Kato

    Multiple Authorship (Only Japanese)

  • Deep Neural Network-Based Speech Recognition with Combination of Speaker-Class Models, Proc. of APSIPA ASC 2015, SP2-2.3, 2015.12

    Tetsuo Kosaka, Kazuki Konno, Masaharu Kato

    Multiple Authorship (Only Japanese)

  • Unsupervised cross-adaptation using language model and deep learning based acoustic model adaptations, Proc. of APSIPA ASC 2014, WA-P-16, 2014.12

    Akira Takagi, Kazuki Konno, Masaharu Kato and Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Elements of difficulty level in Mathematics, International Conference on Software EngineeriProc. of Artificial Intelligence, Networking and Parallel/Distributed Computing , 85-97, 2014.06

    Taku Jiromaru, Tetsuo Kosaka, Tokuro Matsuo

    Multiple Authorship (Only Japanese)

  • Speech recognition with large-scale speaker-class-based acoustic modeling, Proc. of APSIPA ASC 2013, OS.28-SLA.9, 2013.10

    Kazuki Konno, Masaharu Kato and Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • An investigation of vowel substitution rules in the automatic evaluation system of English pronunciation, Proc. of Meetings on Acoustics , 2013.06

    Kei Sato, Masaharu Kato and Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • A time-synchronous histogram equalization for noise robust speech recognition, Proc. of Meetings on Acoustics, 2013.06

    Fumiya Takahashi, Masaharu Kato and Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • Unsupervised Cross-Adaptation Approach for Speech Recognition by Combined Language Model and Acoustic Model Adaptation, Proc. of APSIPA ASC 2011, 2011.10

    Tetsuo Kosaka, Taro Miyamoto, Masaharu Kato

    Multiple Authorship (Only Japanese)

  • Speaker Vector-Based Verification by Phonetic Class-Based Modeling, Journal of Information Assurance and Security, 6(3) 186-194, 2011.03

    Tetsuo Kosaka, Naoki Tadokoro, Masaharu Kato, Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Lecture Speech Recognition Using Discrete-Mixture HMMs, IEEJ Transactions on Electrical and Electronic Engineering, 6(1) 23-29, 2011.01

    Tetsuo Kosaka, Akiyoshi Yamamoto, Takuya Kumakura, Masaharu Kato, Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Unsupervised Speaker Adaptation Using Speaker-Class Models for Lecture Speech Recognition, IEICE Transactions on Information and Systems, E93-D(9) 2363-2369, 2010.09

    Tetsuo Kosaka, Yuui Takeda, Takashi Ito, Masaharu Kato, Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Lecture Speech Recognition by Combining Word Graphs of Various Acoustic Models, Proc. of Interspeech 2010, 2978-2981, 2010.09

    Tetsuo Kosaka, Keisuke Goto, Takashi Ito and Masaharu Kato

    Multiple Authorship (Only Japanese)

  • Speaker Adaptation Based on System Combination Using Speaker-Class Models, Proc. of Interspeech 2010, 546-549, 2010.09

    Tetsuo Kosaka, Takashi Ito, Masaharu Kato, Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Performance Improvement in Automatic Evaluation System of English Pronunciation by Using Various Normalization Methods, Proc. of ICA2010, #257, 2010.08

    Masaru Kusumi, Masaharu Kato, Tetsuo Kosaka, Itaru Matsunaga

    Multiple Authorship (Only Japanese)

  • Speech Recognition in Noise by Using Word Graph Combinations, Proc. of ICA2010, #341, 2010.08

    Shunsuke Kuramata, Masaharu Kato, Tetsuo Kosaka

    Multiple Authorship (Only Japanese)

  • An Algorithm for Fast Calculation of Back-off N-gram Probabilities with Unigram Rescaling, IAENG International Journal of Computer Science, 36(4) 284-288, 2009.11

    Masaharu Kato, Tetsuo Kosaka, Akinori Ito and Shozo Makino

    Multiple Authorship (Only Japanese)

  • Noisy Speech Recognition by using Output Combination of Discrete-Mixture HMMs and Continuous-Mixture HMMs, Proc. of Interspeech 2009, 2379-2382, 2009.09

    Tetsuo Kosaka, You Saito and Masaharu Kato

    Multiple Authorship (Only Japanese)

  • An Improvement of Speaker Vector-Based Speaker Verification, Proc. of International Conference on Information Assurance and Security, 721-724, 2009.08

    Naoki Tadokoro, Tetsuo Kosaka, Masaharu Kato and Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Fast and Robust Training of a Probabilistic Latent Semantic Analysis Model by the Parallel Learning and Data Segmentation, Journal of Communication and Computer, 6(5) 28-35, 2009.05

    Masaharu Kato, Tetsuo Kosaka, Akinori Ito and Shozo Makino

    Multiple Authorship (Only Japanese)

  • Dictation of Japanese speech based on Kana and Kanji character string, International Journal of Computer Processing of Languages, 22(1) 75-98, 2009.03

    AKINORI ITO, HIROAKI KINNO, MASAHARU KATOH, TETSUO KOSAKA and MASAKI KOHDA

    Multiple Authorship (Only Japanese)

  • An Investigation on Speaker Vector-based Speaker Identification under Noisy Conditions, Proc. of International Conference on Audio, Language and Image Processing 2008, 1430-1435, 2008.07

    Yuki Goto, Tatsuya Akatsu, Masaharu Katoh, Tetsuo Kosaka and Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Histogram Equalization for Noise-Robust Speech Recognition Using Discrete-Mixture HMMs, Acoustical Science and Technology, 29(1) 66-73, 2008.01

    Tetsuo Kosaka, Masaharu Katoh, Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Spontaneous Speech Recognition using Discrete-Mixture HMMs, Proc. of ICA2007, CAS-03-034, 2007.09

    Tetsuo Kosaka, Masaharu Katoh and Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Robust Speech Recognition Using Discrete-Mixture HMMs, IEICE Transactions on Information and Systems, E88-D(12) 2811-2818, 2005.12

    T. Kosaka, M. Katoh and M. Kohda

    Multiple Authorship (Only Japanese)

  • Robust Speech Recognition under Non-Stationary Noise Using Discrete-Mixture HMMs, Proc. of NCSP05, 347-350, 2005.03

    T.Kosaka, M.Katoh, M.Kohda

    Multiple Authorship (Only Japanese)

  • Noisy Speech Recognition with Discrete-Mixture HMMs Based on MAP Estimation, Proc. of ICA2004, (2) 1691-1694, 2004.04

    Tetsuo Kosaka, Masaharu Katoh and Masaki Kohda

    Multiple Authorship (Only Japanese)

  • Instantaneous Environment Adaptation Techniques Based on Fast PMC and MAP-CMS Methods, Proc. of ICASSP98, 789-792, 1998.05

    T.Kosaka, H.Yamamoto, M.Yamada, Y.Komori

    Multiple Authorship (Only Japanese)

  • Fast Parallel Model Combination Noise Adaptation Processing, Proc. of Eurospeech97, 1523-1526, 1997.09

    Yasuhiro Komori, Tetsuo Kosaka, Hiroki, Yamamoto, Masayuki Yamada

    Multiple Authorship (Only Japanese)

  • Fast Algorithm for Speech Recognition Using Speaker Cluster HMM, Proc. of Eurospeech97, 2043-2046, 1997.09

    Masayuki Yamada, Yasuhiro Komori, Tetsuo Kosaka, Hiroki Yamamoto

    Multiple Authorship (Only Japanese)

  • Speaker Adaptation for Context-Dependent HMM Using Spatial Relation of Both Phoneme Context Hierarchy and Speakers, Proc. of Eurospeech97, 2039-2042, 1997.09

    Yasuhiro Komori, Tetsuo Kosaka, Masayuki Yamada, Hiroki Yamamoto

    Multiple Authorship (Only Japanese)

  • Fast Speech Recognition Algorithm Under Noisy Environment Using Modified CMS-PMC and Improved IDMM+SQ, Proc. of ICASSP97, 847-850, 1997.04

    H.Yamamoto, T.Kosaka, M.yamada, Y.Komori, M.Fujita

    Multiple Authorship (Only Japanese)

  • Fast Output Probability Computation Using Scalar Quantization and Independent Dimension Multi-Mixture, Proc. of ICASSP96, 893-896, 1996.05

    M.Yamada, H.Yamamoto, T.Kosaka, Y.Komori, Y.Ohora

    Multiple Authorship (Only Japanese)

  • Independent Calculation of Power Parameters on PMC Method, Proc. of ICASSP96, pp. 41-44, 1996.05

    H.Yamamoto, M.Yamada, T.Kosaka, Y.Komori, Y.Ohora

    Multiple Authorship (Only Japanese)

  • Speaker Adaptation Based on Transfer Vector Field Smoothing Using Maximum a Posteriori Probability Estimation, Computer Speech and Language, 10(2) 117-132, 1996.04

    Masahiro Tonomura, Tetsuo Kosaka, Shoichi Matsunaga

    Multiple Authorship (Only Japanese)

  • Speaker-independent speech recognition based on tree-structured speaker clustering, Computer Speech and Language, 10(1) 55-74, 1996.01

    Tetsuo Kosaka, Shoichi Matsunaga, Shigeki Sagayama

    Multiple Authorship (Only Japanese)

  • Speaking-Style and Speaker Adaptation for the Recognition of Spontaneous Dialogue Speech, Proc. of Eurospeech95, 1135-1138, 1995.09

    Shoichi Matsunaga, Tetsuo Kosaka, Tohru Simizu

    Multiple Authorship (Only Japanese)

  • Speaker Adaptation Fitting Training Data Size and Contents, Proc. of Eurospeech95, 1147-1150, 1995.09

    Masahiro Tonomura, Tetsuo Kosaka, Shoichi Matsunaga, Yumi Wakita, Akito Monden

    Multiple Authorship (Only Japanese)

  • Automatic Determination of the Number of Mixture Components for Continuous HMMs Based on a Uniform Variance Criterion, IEICE Tansactions Inf. And Syst., E78-D(6) 642-647, 1995.06

    Tetsuo Kosaka, Shigeki Sagayama

    Multiple Authorship (Only Japanese)

  • Speaker-Independent Phone Modeling Based on Speaker-Dependent HMMs' Composition and Clustering, Proc. of ICASSP95, 441-444, 1995.04

    Tetsuo Kosaka, Shoichi Matsunaga, Mikio Kuraoka

    Multiple Authorship (Only Japanese)

  • Speaker Adaptation Based on Transfer Vector Field Smoothing Using Maximum a Posteriori Probability Estimation, Proc. of ICASSP95, 688-691, 1995.04

    Masahiro Tonomura, Tetsuo Kosaka, Shoichi Matsunaga

    Multiple Authorship (Only Japanese)

  • Tree-Structured Speaker Clustering for Speaker-Independent Continuous Speech Recognition, Proc. of ICSLP94, 1375-1378, 1994.09

    Tetsuo Kosaka, Shoichi Matsunaga, Shigeki Sagayama

    Multiple Authorship (Only Japanese)

  • Tree-Structured Speaker Clustering for Fast Speaker Adaptation, Proc. of ICASSP94, 245-248, 1994.04

    Tetsuo Kosaka, Shigeki Sagayama

    Multiple Authorship (Only Japanese)

  • A Dynamic Approach to Speaker Adaptation of Hidden Markov Networks for Speech Recognition, Proc. of Eurospeech93, 363-366, 1993.09

    Tetsuo Kosaka, Edward Willems, Jun-Ichi Takami, Shigeki Sagayama

    Multiple Authorship (Including Foreigners)

  • Rapid Speaker Adaptation Using Speaker-Mixtue Allophone Models Applied to Speaker-Independent Speech Recognition, Proc. of ICASSP93, 570-573, 1993.04

    Tetsuo Kosaka, Jun-Ichi Takami, Shigeki Sagayama

    Multiple Authorship (Only Japanese)

  • An Algorithm for Automatic HMM Structure Generation in Speech Recognition, Proc. of Fourth Australian International Conference on Speech Science and Technology , 104-109, 1992.12

    Tetsuo Kosaka, Shigeki Sagayama

    Multiple Authorship (Only Japanese)

  • ATREUS: Continuous Speech Recognition Systems at ATR Interpreting Telephony Research Laboratiories, Proc. of Fourth Australian International Conference on Speech Science and Technology, 324-329, 1992.12

    S. Sagayama, M. Sugiyama, K. Ohkura, J. Takami, A. Nagai, H. Singer, H. Hattori, K. Fukuzawa, Y. Kato, K. Yamaguti, T. Kosaka, A. Kurematsu

    Multiple Authorship (Only Japanese)

display all >>

Books

  • Acoustics, Speech and Human-Machine Communication (Shuichi ITAHASHI et al. ed.) Chapter 2.3, No. 20., AOSOSHA Publishing company, 2013.12

    Tetsuo Kosaka et al.

  • E-Activity and Intelligent Web Construction: Effects of Social Design, Chapter16, IGI Global, 2011.05

    Tetsuo Kosaka, Takashi Kusama, Masaharu Kato, Masaki Kohda

  • Speech Recognition, I-Tech Education and Publishing, 2008.11

    France Mihelic and Janez Zibert (Ed)

  • Robust Speech Recognition and Understanding, I-Tech Education and Publishing, 2007.06

    M. Grimm and K. Kroschel (Ed)

Grant-in-Aid for Scientific Research

  • Grant-in-Aid for Scientific Research(C),2022.04 - 2025.03

  • Grant-in-Aid for Scientific Research(C),2019.04 - 2022.03

  • Grant-in-Aid for Scientific Research(C),2016.04 - 2019.03

  • Grant-in-Aid for Scientific Research(C),2013.04 - 2016.03

  • Grant-in-Aid for Scientific Research(C),2010.04 - 2013.03

  • Grant-in-Aid for Scientific Research(C),2004.04 - 2007.03

display all >>