Tomoko MATSUI

Professor, Intelligent Information Processing Group, Department of Statistical Modeling

 

Phone :

+81-50-5533-8537

Fax(shared):

+81-42-526-4335

E-mail:

tmatsuihttp://www.ism.ac.jp/souran/icon/atISM.gif

Personal
Homepage:

http://www.ism.ac.jp/~tmatsui/

Education:

BS degree, Department of Information Science, Tokyo Institute of Technology, April 1984 - March 1986.
MS degree, Department of Information Science, Tokyo Institute of Technology, April 1986 - March 1988.

Degree:

Ph. D., Tokyo Institute of Technology, 1997, Title of Ph. D. Thesis: A Basic Study of a High-Performance Speaker Recognition System.

Employment:

1988. 4 - 1998.10  Researcher, NTT Human Interface Laboratories
1998.11 - 2000. 3  Researcher, ATR Interpreting Telecommunications Research Laboratories
2000. 4 - 2002.12  Senior Researcher, ATR Spoken Language Translation Research Laboratories
2000 - 2002           Part-time Teacher, Kansai University
2001. 1 - 2001. 6   Visiting Researcher, Bell Laboratories, Lucent Technologies
2003. 1 - 2005. 3   Associate Professor, the Institute of Statistical Mathematics

2004. 1 - 2004. 3   Visiting Researcher, Georgia Institute of Technology
2005. 4 - 2008. 3        Associate Professor, Department of Statistical Modeling, the Institute of Statistical Mathematics

2007. 1 - 2007. 3   Visiting Researcher, University of British Columbia

2008. 4 - now         Professor, Department of Statistical Modeling, the Institute of Statistical Mathematics
2009. 1 - 2009. 2   Visiting Professor, Max Planck Institute for Biological Cybernetics
2010. 2    Visiting Professor, Mines ParisTech

Honors &
Awards:

Paper Award of the IEICE (1993)

Research
Fields:

Speech Processing, Pattern Recognition, Machine Learning

Research
Themes:

Statistical Modeling for Multimodal Processing

Selected Papers
[Papers]
(1)Birkenes, O., Matsui, T., Tanabe, K., Siniscalchi, S. M., Myrvoll, T. A., and Johnsen, M. H. (2010): Penalized Logistic Regression with HMM Log-Likelihood Regressors for Speech Recognition, IEEE Trans. SAP (to appear).
(2)Yamada, M., Sugiyama, M., Matsui, T., (2010): Semi-supervised Speaker Identification under Covariate Shift, Signal Processing (to appear).
(3)Lane, I. R., Kawahara, T., Matsui, T. and Nakamura, S. (2007): Out-of-domain utterance detection using classification confidences of multiple topics, IEEE Trans. Speech & Audio Process., 15(1),150-161.
(4)Matsui, T. and Tanabe, K.(2006): Comparative Study of Speaker Identification Methods: dPLRM, SVM and GMM, IEICE, E89-D(3), 1066-1073.
(5)Matsui, T., Soong, F.K. and Juang, B. -H.(2005):Verification of Multi-Class Recognition Decision: A Classification Approach, IEICE, E88-D(3), 455-462.
(6)Lane, I. R., Kawahara, T. Matsui, T. and Nakamura, S.(2005): Dialogue Speech Recognition by Combining Hierarchical Topic Classification and Language Model Switching, IEICE, E88-D(3), 446-454.
(7)Jitsuhiro, T., Matsui, T. and Nakamura, S.(2004): Automatic generation of non-uniform HMM topologies based on the MDL criterion, IEICE, E87-D(8), 2121-2129.
(8)Matsui, T. and Aikawa, K.(2003): Robust model for speaker verification against session-dependent utterance variation, IEICE, E86-D(4) , 712-718.
(9)Markov, K., Matsui, T., Gruhn, R. Zhang, J. and Nakamura, S.(2003): Noise and channel distortion robust ASR system for DARPA SPINE2 task, IEICE, E86-D(3), 497-504.
(10)Zhang, J., Markov, K., Matsui, T. and Nakamura, S. (2003): A study on acoustic modeling of pauses for recognizing noisy conversational speech, IEICE, E86-D(3), 489-496.
(11)Okuda, K., Matsui, T. and Nakamura, S. (2003): An acoustic modeling for stressed speech in error recovery Japanese utterances (in Japanese), IEICE, J86-D-II(1), 42-51.
(12)Okuda, K., Matsui, T., Naito, M., Sagisaka, Y. and Nakamura, S. (2002): Creation and evaluation of a large-scale Japanese speech database with wide regional and age distribution (in Japanese), The Journal of The Acoustical Society of Japan, 58(9) , 569-578.
(13)Matsui, T. and Furui, S. (1998): N-best-based unsupervised speaker adaptation for speech recognition,Computer Speech and Language, 12, 41-50.
(14)Matsui, T., Nishitani, T. and Furui, S. (1998): A study of model and a priori threshold updating in speaker verification (in Japanese), IEICE, J81-D-II(2), 268-276.
(15)Matsui, T., Kanno, T. and Furui, S. (1996): Speaker recognition using HMM composition in noisy environments, Computer Speech and Language, 10, 107-116.
(16)Matsui, T. and Furui, S. (1996): Text-prompted speaker recognition (in Japanese), IEICE, J79-D-II(5), 647-656.
(17)Matsui, T. and Furui, S. (1996): Likelihood normalization using a phoneme- and speaker-independent model for speaker verification, Speech Communication, 17(1-2), 109-116.
(18)Matsui, T. and Furui, S. (1994): Comparison of text-independent speaker recognition methods using VQ-distortion and discrete/continuous HMMs (in Japanese), IEICE, J77-A(4), 601-606.
(19)Matsui, T. and Furui, S. (1992): Text-independent speaker recognition using vocal tract and pitch information (in Japanese), IEICE, J75-A(4), 703-709.
[Letter]
(1)Matsui, T. and Furui, S. (1994): Comparison of text-independent speaker recognition methods using VQ-distortion and discrete/continuous HMMs, IEEE Trans., SAP, 2, 456-459.

Membership of Academic Societies
Member of IEEE, ISCA, the Acoustic Society of Japan, and IEICE

Professional Services
Speech Communication Subject Editor, ISCA Workshop Odyssey Scientific Committee, IEICE Councilor, IEICE IBIS Technical Group Member, IEICE BS Technical Group Member, SC37 WG3 Member of Information Technology Standards Commission of Japan, and Reviewing Staff of ASJ and IEICE Journals

HOME