Unsupervised HMM adaptation based on speech-silence discriminati

Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

704233, G10L 506

Patent

active

060760575

ABSTRACT:
An unsupervised, discriminative, sentence level, HMM adaptation based on speech-silence classification is presented. Silence and speech regions are determined either using a speech end-pointer or the segmentation obtained from the recognizer in a first pass. The discriminative training procedure using a GPD or any other discriminative training algorithm, employed in conjunction with the HMM-based recognizer, is then used to increase the discrimination between silence and speech.

REFERENCES:
patent: 4481593 (1984-11-01), Bahler
patent: 4741036 (1988-04-01), Bahl et al.
patent: 5003601 (1991-03-01), Watari et al.
patent: 5333275 (1994-07-01), Wheatley et al.
patent: 5579436 (1996-11-01), Chou et al.
patent: 5606644 (1997-02-01), Chou et al.
patent: 5617486 (1997-04-01), Chow et al.
patent: 5649057 (1997-07-01), Lee et al.
patent: 5710864 (1998-01-01), Juang et al.
patent: 5715367 (1998-02-01), Gillick et al.
patent: 5717826 (1998-02-01), Setlur
patent: 5727124 (1998-03-01), Lee et al.
patent: 5737486 (1998-04-01), Iso
patent: 5778340 (1998-07-01), Hattori
patent: 5778341 (1998-07-01), Zeljkovic
patent: 5778342 (1998-07-01), Erell et al.
Wilpon et al. Automatic Recognition of Keywords in Unconstrained Speech Using Hidden Markov Models. IEEE Transaction on Signal Processing. pp. 1870-1878, Nov. 1990.
Mikkilineni et al. Discriminative Training of A Connected Digit Recogizer with Fixed Filler Models and its Applciation to Telephone Network Service Systmes. IEEE, May 1996.
De Souza. A Statistical Approach to the Design of an Adapative Self-Normalizing Silence Detector. IEEE Transactions on Acoustics, Speech and Signal Processing. No. 3, Jun. 1993.
de Veth et al. Limited Paramter Hidden Markov Models for Connected Digit Speaker Verfication Over Telephone Channels. IEEE, Apr. 1993.
McDonough et al. An Approach to Speaker Adaption based on Analytic Functions. IEEE. 721-724, Aug. 1994.
Matsui et al. N-Best-Based Instantaneous Speaker Adaption Method for Speech Recognition. 973-976, 1996.
Leggetter et al. Maximum likelhood linear regression for speaker adaptation of continuous density hiddent markov Models. 171-185, 1995.
Cox et al. Unsupervised Speaker Adaptation by Probabilistic Spectrum Fitting. IEEE. 1989.
Chou et al. Minimum error rate Training Based on N-Best String Models. IEEE. 652-655, 1993.
Lawrence R. Rabiner et al., "A Segmental k-Means Training Procedure for Connected Word Recognition", AT&T Technical Journal, vol. 65, Issue 3, May/Jun./ 1986, pp. 21-31.
W. Chou et al., "Minimum Error Rate Training Based on N-Best String Models", IEEE, 1993, pp. II-652-II-655.
Lawrence Rabiner et al., Fundamentals of Speech Recognition, Prentice Hall, pp. pp. 69-139 and 200-209.

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Unsupervised HMM adaptation based on speech-silence discriminati 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 HMM adaptation based on speech-silence discriminati, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Unsupervised HMM adaptation based on speech-silence discriminati will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2078110

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.