Speaker verification method and apparatus using mixture decompos

Details Speaker verification method and apparatus using mixture decompos Speaker verification method and apparatus using mixture decompos

1995-05-22

1997-11-11

Zele, Krista M.

395 256, G10L 506

Patent

active

056872879

ABSTRACT:
A new speaker verification method, termed Mixture Decomposition Discrimination (MDD), and a new apparatus for using MDD are presented. MDD takes mixture component score information from a speaker independent recognizer and transmits this information while it is still decomposed as a mixture of component scores that indicate the response of the states of a Hidden Markov Model (HMM) before this information is combined into a single speaker independent recognizer parameter. MDD can be very effective in improving the performance of existing verification methods based on speaker dependent HMMs with cohort normalization because the errors of the two speaker verification methods are very uncorrelated statistically. Experimental results have shown that when MDD is incorporated into a system that also uses speaker dependent HMMs, the resulting hybrid system has its average equal error rate reduced by 46% compared to cohort normalized speaker independent HMMs. MDD is used with a speaker dependent linear discriminator which has relatively low computational and storage requirements. Thus, the increased performance of a hybrid MDD-Cohort Normalized HMM system may be achieved with minimal increase in computational and data storage assets.

REFERENCES:
A. E. Rosenberg et al., "Connected Word Talker Verification Using Whole Word Hidden Markov Models", Proc. 1991 IEEE Int. Conf. on Acoust., Speech and Signal Processing, May 1991, pp. 381-384.
A. E. Rosenberg et al., "The Use of Cohort Normalized Scores for Speaker Verification", Proc. 1992 Int. Conf. on Spoken Language Processing, Oct. 1992, pp. 599-602.
C. Liu et al., "Speaker Recognition Based on Minimum Error Discriminative Training", Proc. 1994 IEEE Int. Conf. on Acoust., Speech and Signal Processing, vol. 1, Apr. 1994, pp. 325-328.
J. M. Naik et al., "A Hybrid HMM-MLP Speaker Verification Algorithm for Telephone Speech", Proc. 1994 IEEE Int. Conf. on Acoust., Speech and Signal Processing, vol. 1, Apr. 1994, pp. 153-156.
K. R. Farrell et al., "Speaker Identification Using Neural Tree Networks", Proc. 1994 IEEE Int. Conf. on Acoust., Speech and Signal Processing, vol. 1, Apr. 1994, pp. 165-168,
J. Sorensen et al., "Hierarchical Pattern Classification for High Performance Text-Independent Speaker Verification Systems", Proc. 1994 IEEE Int. Conf. on Acoust., Speech and Signal Processing, vol. 1, Apr. 1994, pp. 157-160.
L. P. Netsch et al., "Speaker Verification Using Temporal Decorrelation Post-Processing", Proc. 1992 IEEE Int. Conf. on Acoust., Speech and Signal Processing, vol. 1, Mar. 1992, pp. 181-184.
T. E. Jacobs et al., "A Field Study of Performance Improvements in HMM-based Speaker Verification", Proc. 1994 IEEE IVTTA Workshop, Sep. 1994, pp. 121-124.
J. G. Wilpon et al., "A Modified K-Means Clustering Algorithm for Use in Isolated Work Recognition", IEEE Trans. on Acoustics, Speech and Signal Processing, vol. 33, Jun. 1985, pp. 587-594.
A. Higgins et al., "Speaker Verification Using Randomized Phrase Prompting", Digital Signal Processing, Mar. 1991, pp. 89-106.

Affiliated with

Also associated with

Rating

Speaker verification method and apparatus using mixture decompos does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Speaker verification method and apparatus using mixture decompos, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speaker verification method and apparatus using mixture decompos will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-1235614