Vector-field-smoothed Bayesian learning for fast and incremental speaker/telephone-channel adaptation

Computer Speech and Language

Volumn 11, Issue 2, 1997, Pages 127-146

  Takahashi, Jun Ichi ^a   Sagayama, Shigeki ^a  

^a NTT CORPORATION   (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

COMMUNICATION CHANNELS (INFORMATION THEORY); DATA PROCESSING; ERROR CORRECTION; LEARNING SYSTEMS; SPEECH TRANSMISSION; VOCABULARY CONTROL;

BAYESIAN LEARNING; INTRA-CLASS TRAINING; VECTOR FIELD SMOOTHING (VFS);

SPEECH RECOGNITION;

EID: 0031118076     PISSN: 08852308     EISSN: None     Source Type: Journal    
DOI: 10.1006/csla.1996.0025     Document Type: Article

References (19)

1. Acero, A. & Stern, R. M. (1990). Environmental robustness in automatic speech recognition. In Proceedings of the IEEE 1990 International Conference on Acoustics, Speech and Signal Processing, Albuquerque, NM, pp. 849-852.
2. Gauvain, J.-L. & Lee, C.-H. (1991). Bayesian learning of Gaussian mixture densities for hidden Markov models. In Proceedings of the DARPA Speech and Natural Language Workshop, Arden House, pp. 272-277.
3. Gauvain, J.-L. & Lee, C.-H. (1994). Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chains. IEEE Transactions on Acoustics, Speech and Signal Processing 2, 291-298.
4. Lee, C.-H. & Gauvain, J.-L. (1993). Speaker adaptation based on MAP estimation of HMM parameters. In Proceedings of the IEEE 1993 International Conference on Acoustics, Speech and Signal Processing, Minneapolis, MN, pp. II-558-561.
5. Lee, C.-H., Lin, C. H. & Juang, B. H. (1991). A study on speaker adaptation of the parameters of continuous density hidden Markov models. IEEE Transactions on Acoustics, Speech and Signal Processing 39, 806-814.
6. Liu, F. H., Stern, R. M., Acero, A. & Moreno, P. J. (1994). Environment normalization for robust speech recognition using direct cepstral comparison. In Proceedings of the IEEE 1994 International Conference on Acoustics, Speech and Signal Processing, Minneapolis, MN, pp. 11-61-64.
7. Markel, J. D. & Gray, A. H. (1976). Linear Prediction of Speech, pp. 146-151. Springer-Verlag, Berlin.
8. Matsuoka, T. & Lee, C.-H. (1993). A study of on-line Bayesian adaptation for HMM-based speech recognition. In Proceedings of the European Conference on Speech Communication and Technology, Berlin, Germany, pp. 815-818.
9. Ohkura, K., Sugiyama, M. & Sagayama, S. (1992). Speaker adaptation based on transfer vector field smoothing with continuous mixture density HMMs. In Proceedings of the 1992 International Conference on Spoken Language Processing, Banff, Canada, pp. 369-372.
10. Takahashi, J. & Sagayama, S. (1994a). Telephone line characteristic adaptation using vector field smoothing technique. In Proceedings of the 1994 International Conference of Spoken Language Processing, Yokohama, Japan, pp. 991-994.
11. Takahashi, J. & Sagayama, S. (1994b). Fast telephone channel adaptation based on vector field smoothing technique. In Proceedings of the IEEE 2nd Workshop on Interactive Voice Technology for Telecommunications Applications, Kyoto, Japan, pp. 97-100.
12. Takahashi, J. & Sagayama, S. (1994c). Vector-field-smoothed Bayesian learning for fast speaker adaptation. In Proceedings of the 1994 Fall Meeting, Acoustical Society of Japan, 2-8-19, Kumamoto, Japan, pp. 75-76 (in Japanese).
13. Takahashi, J. & Sagayama, S. (1994d). Vector-field-smoothed Bayesian learning for incremental speaker/telephone-line adaptation. Technical Report of the Institute of Electronics, Information and Communication Engineers, SP-94-74, Japan, pp. 33-40 (in Japanese).
14. Takahashi, J. & Sagayama, S. (1995a). Vector-field-smoothed Bayesian learning for incremental speaker adaptation. In Proceedings of the IEEE 1995 International Conference on Acoustics, Speech and Signal Processing, Detroit, MI, U.S.A., pp. 696-699.
15. Takahashi, J. & Sagayama, S. (19956). Four-level tied-structure for efficient representation of acoustic modeling. In Proceedings of the IEEE 1995 International Conference on Acoustics, Speech and Signal Processing, Detroit, MI, pp. 520-523.
16. Tonomura, M., Kosaka, T. & Matsunaga, S. (1994a). Unified speaker adaptation algorithm of maximum a posteriori estimation and transfer vector field smoothing. Technical Report of the Institute of Electronics, Information and Communication Engineers, SP94-51, Japan, pp. 25-30 (in Japanese)
17. Tonomura, M., Kosaka, T. & Matsunaga, S. (19946). Speaker adaptation based on transfer vector field smoothing model using maximum a posteriori probability estimation. In Proceedings of the 1994 Fall Meeting, Acoustical Society of Japan, 2-8-20, Japan, pp. 77-78 (in Japanese).
18. Tonomura, M., Kosaka, T. & Matsunaga, S. (1995). Speaker adaptation based on transfer vector field smoothing using maximum a posteriori probability estimation. In Proceedings of the IEEE 1995 International Conference on Acoustics, Speech and Signal Processing, Detroit, MI, pp. 688-691.
19. Tsurumi, Y. & Nakagawa, S. (1994). An unsupervised speaker adaptation method for continuous parameter HMM by maximum a posteriori probability estimation. In Proceedings of the 1994 International Conference on Spoken Language Processing, Yokohama. Japan, pp. 431-434.