Speaker-independent speech recognition based on tree-structured speaker clustering

Computer Speech and Language

Volumn 10, Issue 1, 1996, Pages 55-74

  Kosaka, Tetsuo ^a   Matsunaga, Shoichi ^a   Sagayama, Shigeki ^b  

^a ADVANCED TELECOMMUNICATIONS RESEARCH INSTITUTE INTERNATIONAL   (Japan)

^b NTT CORPORATION   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

DATA STRUCTURES; MARKOV PROCESSES; PATTERN RECOGNITION SYSTEMS; PROBLEM SOLVING; SPEECH; SPEECH ANALYSIS; SPEECH COMMUNICATION;

ERROR RATE; HIDDEN MARKOV NETWORK; SPEAKER CLUSTERING; SPEAKER INDEPENDENT; UNSUPERVISED SPEAKER ADAPTATION;

SPEECH RECOGNITION;

EID: 0029735634     PISSN: 08852308     EISSN: None     Source Type: Journal    
DOI: 10.1006/csla.1996.0004     Document Type: Article

References (17)

1. Chow, Y. L., Dunham, M. O., Kimball, O. A., Krasner, M. A., Kubala, G. F., Makhoul, J., Price, P. J., Roucos, S. & Schwartz, R. M. (1987). BYBLOS: the BBN continuous speech recognition system. Proceedings of ICASSP'87, pp. 89-92.
2. Fukunaga, K. (1990). Introduction to Statistical Pattern Recognition (second edition). Academic Press, San Diego, CA.
3. Gauvain, J.-L. & Lee, C.-H. (1993). Speaker adaptation based on map estimation of HMM parameters. Proceedings of ICASSP'93, pp. 558-561.
4. Juang, B.-H. & Rabiner, L. R. (1985). A probabilistic distance measure for hidden Markov models. AT&T Technical Journal 2, 391-408.
5. Kosaka, T., Matsunaga, S. & Sagayama, S. (1995). Tree-structured speaker clustering for speaker adaptation. IEICE Transactions D-II 1, 1-9 (in Japanese).
6. Kosaka, T. & Sagayama, S. (1994). Tree-structured speaker clustering for fast speaker adaptation. Proceedings of ICASSP'94, pp. 245-248.
7. Kosaka, T., Takami, J. & Sagayama, S. (1993). Rapid speaker adaptation using speaker-mixture allophone models applied to speaker-independent speech recognition. Proceedings of ICASSP'93, pp. 570-573.
8. Lee, K.-F. (1990). Context-dependent phonetic hidden Markov models for speaker-independent continuous speech recognition. IEEE Transactions on Acoustics, Speech and Signal Processing 4, 599-609.
9. Linde, Y., Buzo, A. & Gray, R. M. (1980). An algorithm for vector quantizer design. IEEE Transactions on Communication 1, 84-95.
10. Mathan, L. & Miclet, L. (1990). Speaker hierarchical clustering for improving speaker-independent HMM word recognition. Proceedings of ICASSP'90, pp. 149-152.
11. Nagai, A., Sagayama, S. & Kita, K. (1991). Phoneme-context-dependent LR parsing algorithms for HMM-based continuous speech recognition. Proceedings of Eurospeech'91, pp. 1397-1400.
12. Niyogi, P. & Zue, V. W. (1991). Correlation analysis of vowels and their application to speech recognition. Proceedings of Eurospeech'91, pp. 1253-1256.
13. Ohkura, K., Sugiyama, M. & Sagayama, S. (1992). Speaker adaptation based on transfer vector field smoothing with continuous mixture density HMMs. Proceedings of ICSLP'92, pp. 369-372.
14. Sugamura, N., Shikano, K. & Furui, S. (1983). Isolated word recognition using phoneme-like templates. Proceedings of ICASSP'83, pp. 723-726.
15. Sugiyama, M. (1986). Unsupervised speaker adaptation methods for vowel templates. Proceedings of ICASSP'86, pp. 2639-2642.
16. Takami, J., Kosaka, K. & Sagayama, S. (1992). Automatic generation of speaker-common hidden Markov network by adding the speaker splitting domain to the successive state splitting algorithm. Proceedings of Spring Meeting, Acoustic Society of Japan, pp. 155-156 (in Japanese).
17. Takami, J. & Sagayama, S. (1992). A successive state splitting algorithm for efficient allophone modeling. Proceedings of ICASSP'92, pp. 573-576.