Self-learning speaker adaptation based on spectral bias source d

Details Self-learning speaker adaptation based on spectral bias source d Self-learning speaker adaptation based on spectral bias source d

1996-09-12

1998-08-11

Hudspeth, David R.

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

704240, 704242, 704254, G10L 506, G10L 708

Patent

active

057941920

ABSTRACT:
A speaker adaptation technique based on the separation of speech spectra variation sources is developed for improving speaker-independent continuous speech recognition. The variation sources include speaker acoustic characteristics, and contextual dependency of allophones. Statistical methods are formulated to normalize speech spectra based on speaker acoustic characteristics and then adapt mixture Gaussian density phone models based on speaker phonologic characteristics. Adaptation experiments using short calibration speech (5 sec./speaker) have shown substantial performance improvement over the baseline recognition system.

REFERENCES:
patent: 4363102 (1982-12-01), Holmgren et al.
patent: 4903305 (1990-02-01), Gillick et al.
patent: 5033087 (1991-07-01), Bahl et al.
Kubala, Francis et al. "Speaker Adaptation From a Speaker Independent Training Corpus," IEEE ICASSP, pp. 137-140, Apr. 1990.
Huang, X.D. and Lee, K.F., "On Speaker-Independent, Speaker-Dependent, and Speaker-Adaptive Speech Recognition," IEEE ICASSP, pp. 877-868, May 1991.
Rozzi, William A. and Stern, Richard M., "Speaker Adaptation in continuous Speech Recognition Via Estimation of Correlated Mean Vectors," IEEE ICASSP, pp. 865-868, May 1991.
Schmidbauer, O., Tebelskis, J., "An LVQ Based Reference Model for Speaker Adaptive Speech Recognition," IEEE ICASSP, pp. I-441-I444, Mar. 1992.
Furui, Sadaoki, "Unsupervised Speaker Adaptation Method Based on Hierarchical Spectral Clustering," ICASSP, pp. 286-289, May 1989.
Hunt, Melvyn, "Session S. Speech Communication III: Speech Recognition," J. Acoust. Soc. Am. Suppl. 1, vol. 69, Spring 1981, pp. S41-S42.
Matsumoto, Hiroshi, Wakita, Hisashi, "Vowel Normalization by frequency Warped Spectral Matching," Elsevier Science Publ. B.V., Speech Communication, vo. 5, 1986, pp. 239-251.
Cox, S.J., Bridle, J.S., "Unsupervised Speaker Adaptation by Probabilistic Spectrum Fitting," ICASSP, pp. 294-297, May 1989.
Cox, S.J., Bridle, J.S. "Simultaneous Speaker Normalization and Utterance Labelling Using Bayesian/Neural Net Techniques," IEEE ICASSP, pp. 161-164,Apr. 1990.
Lee, Chin-Hui et al., "A Study on Speaker Adaptation of Continuous Density HMM Parameters," IEEE ICASSP, pp. 145-148, Apr. 1990.
S. Furui, "Unsupervised Speaker Adaptation Method Based on Hierarchical Spectral Clustering", Proc. ICASSP, pp. 286-289, Glasgow, Scotland, May 1989.
Y. Zhao, H. Wakita, and X. Zhwang, An HMM Based Speaker-Independent Continuous Speech Recognition System, Proc. ICASSP, pp. 333-336, Toronto, Canada, May, 1991.
K.F. Lee, "Large Vocabulary Speaker-Independent Continuous Speech Recognition: The SPHINX System" PhD Dissertation, Carnegie Mellon Univ, CMU-CS-88-148, April, 1988, pp. 19-42.

Affiliated with

Also associated with

Rating

Self-learning speaker adaptation based on spectral bias source d does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Self-learning speaker adaptation based on spectral bias source d, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Self-learning speaker adaptation based on spectral bias source d will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-403558