Cepstral correction vector quantizer for speech recognition

Details Cepstral correction vector quantizer for speech recognition Cepstral correction vector quantizer for speech recognition

1994-09-30

1997-01-28

Tung, Kee M.

395 231, 395 239, 395 26, G10L 302

Patent

active

055985050

ABSTRACT:
A method for correcting cepstral vectors representative of speech generated in a test environment by use of a vector quantization (VQ) system with a codebook of vectors that was generated using speech and acoustic data from a different (training) environment. The method uses a two-step correction to produce test environment cepstral vectors with reduced non-speech acoustic content. The first correction step subtracts, from the test vector, a coarse correction vector that is computed from an average of test environment cepstral vectors. The second step involves a VQ of the coarsely corrected test vector at each node of the VQ tree. The third step is the addition of a fine correction vector to the coarsely corrected test vector that is generated by subtracting a running (moving) average of the coarsely corrected test vectors associated with the deepest VQ tree node from the VQ vector closest to the coarsely corrected test vector. The method is independent of the means used to generate the cepstral vectors and the corrected output cepstra vectors may be used in various speech processing and classifying systems. The method is adaptable to non-stationary environments.

REFERENCES:
patent: 5010574 (1991-04-01), Wang
patent: 5208862 (1993-05-01), Ozawa
patent: 5396625 (1995-03-01), Parkes
K. F. Lee, Automatic Speech Recognition, The Development of the SPHINX System, Kluwer Academic Publishers, Boston, 1989.
Alejandro Acero et al., "Robust Speech Recognition by Normalization of the Acoustic Space," ICASSP 91, Speech Processing 2, pp. 893-896 (May 14-17, 1991).
Fu-Hua Liu et al., "Efficient Joint Compensation of Speech for the Effects of Additive Noise and Linear Filtering," ICASSP92, Speech Processing 1, pp. I-257 through I-260 (Mar. 23-26, 1992).
Francis T. Connolly et al., "Fast Algorithms for Complex Matrix Multiplication Using Surrogates," IEEE Trans. on Acoustics, Spch & Signal Processing, vol. 37, No. 6, pp. 938-949 (Jun. 1989).
Hynek Hermansky, "Perceptual Linear Predictive (PLP) Analysis of Speech," The Journal of the Acoustical Society of America, vol. 87, No. 4, pp. 1738-1752 (Apr. 1990).
Hynek Hermansky et al., "Recognition of Speech in Additive and Convolutional Noise Based on RASTA Spectral Processing," ICASSP-93, Spch Processing, pp. II-83-II-86 (Apr. 27-30, 1993).
H. Sakoe et al., "Dynamic Programming Algorithm Optimization for Spoken Word Recognition," IEEE Trans. on Acoustics, Spch, & Signal Processing, vol. 26, No. 1, pp. 43-49 (Feb. 1978).
Alejandro Acero et al., "Environmental Robustness in Automatic Speech Recognition," ICASSP 90, vol. 2, Speech Processing 2, pp. 849-852 (Apr. 3-6,1990).
J. D. Markel et al., Linear Prediction of Speech, Springer-Verlag, Berlin, pp. 227-233 (1976).
Yoseph Linde et al., "An Algorithm for Vector Quantizer Design," IEEE Transactions on Communications, vol. COM-28, No. 1, pp. 84-95 (Jan. 1980).
L. R. Rabiner et al., "Note on the Properties of a Vector Quantizer for LPC Coefficients," The Bell System Technical Journal, vol. 62, No. 8, Part 1, pp. 2603-2616 (Oct. 1983).
L. R. Bahl et al., "Large Vocabulary Natural Language Continuous Speech Recognition,"ICASSP.89, vol. 1, Speech Processing 1, pp. 465-467 (May 23-26, 1989).
Robert M. Grey, "Vector Quantization," IEEE ASSP Magazine, vol. 1, No. 2, pp. 4-29 (Apr. 1984).
G. S. Kang et al., "Quality Imprvmnt of LPC-Processed Noisy Spch by Using Spectral Subt.," IEEE Trans. on Acoustics, Spch & Signal Proc., vol. 37, No. 6, pp. 939-942 (Jun. 1989).

Affiliated with

Also associated with

Rating

Cepstral correction vector quantizer for speech recognition does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Cepstral correction vector quantizer for speech recognition, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Cepstral correction vector quantizer for speech recognition will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-946637