Unsupervised incremental adaptation using maximum likelihood...

Details Unsupervised incremental adaptation using maximum likelihood... Unsupervised incremental adaptation using maximum likelihood...

2007-09-11

2007-09-11

Opsasnick, Michael N. (Department: 2626)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

C704S222000, C704S244000

Reexamination Certificate

active

11215415

ABSTRACT:
In a speech recognition system, a method of transforming speech feature vectors associated with speech data provided to the speech recognition system includes the steps of receiving likelihood of utterance information corresponding to a previous feature vector transformation, estimating one or more transformation parameters based, at least in part, on the likelihood of utterance information corresponding to a previous feature vector transformation, and transforming a current feature vector based on maximum likelihood criteria and/or the estimated transformation parameters, the transformation being performed in a linear spectral domain. The step of estimating the one or more transformation parameters includes the step of estimating convolutional noise Niαand additive noise Niβfor each ith component of a speech vector corresponding to the speech data provided to the speech recognition system.

REFERENCES:
patent: 4817156 (1989-03-01), Bahl et al.
patent: 5230037 (1993-07-01), Giustiniani et al.
patent: 5625749 (1997-04-01), Goldenthal et al.
patent: 5937384 (1999-08-01), Huang et al.
patent: 6006175 (1999-12-01), Holzrichter
patent: 6188982 (2001-02-01), Chiang
patent: 6202047 (2001-03-01), Ephraim et al.
patent: 6236962 (2001-05-01), Kosaka et al.
patent: 6381569 (2002-04-01), Sih et al.
patent: 6389393 (2002-05-01), Gong
patent: 6539352 (2003-03-01), Sharma et al.
patent: 6625587 (2003-09-01), Erten et al.
patent: 6760701 (2004-07-01), Sharma et al.
Ephraim, Y.; Rahim, M.; “On second-order statistics and linear estimation of cepstral coefficients”,□□Speech and Audio Processing, IEEE Transactions on□□vol. 7, Issue 2, Mar. 1999 pp. 162-176 □□.
C.J. Leggetter et al. entitled “Speaker Adaptation of Continuous Denisty HMMs Using Multivariate Linear Regression,”International Conference on Spoken Language Processing, pp. 451-454 (1994).
D. Yuk et al. entitled “Adaptation to Environment and Speaker Using Maximum Likelihood Neural Networks,”Eurospeech, pp. 2531-2534 (Sep. 1999).
L.E. Baum entitled “An Inequality and Associated Maximization Technique in Statistical Estimation for Probabilistic Functions of Markov Processes,”Inequalitites, Chapter 3, pp. 1-8 (1972).
G.D. Forney, Jr. entitled “The Viterbi Algorithm,”Proceedings of the IEEE, vol. 61, No. 3, pp. 268-278 (Mar. 1973).
S. Furui entitled “Cepstral Analysis Technique for Automatic Speaker Verification,”IEEE Transactions on Acoustics, Speech and Signal Processing, ASSP-29(2), pp. 254-272, Sep. 1996.

Affiliated with

Also associated with

Rating

Unsupervised incremental adaptation using maximum likelihood... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Unsupervised incremental adaptation using maximum likelihood..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Unsupervised incremental adaptation using maximum likelihood... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3798784