Training speech recognition word models from word samples...

Details Training speech recognition word models from word samples... Training speech recognition word models from word samples...

2006-11-07

2006-11-07

Azad, Abul K. (Department: 2626)

Data processing: speech signal processing, linguistics, language

Speech signal processing

Recognition

C704S254000

Reexamination Certificate

active

07133827

ABSTRACT:
A new word model is trained from synthetic word samples derived by Monte Carlo techniques from one or more prior word models. The prior word model can be a phonetic word model and the new word model can be a non-phonetic, whole-word, word model. The prior word model can be trained from data that has undergone a first channel normalization and the synthesized word samples from which the new word model is trained can undergo a different channel normalization similar to that to be used in a given speech recognition context. The prior word model can have a first model structure and the new word model can have a second, different, model structure. These differences in model structure can include, for example, differences of model topology; differences of model complexity; and differences in the type of basis function used in a description of such probability distributions.

REFERENCES:
patent: 3770892 (1973-11-01), Clapper
patent: 4481593 (1984-11-01), Bahler
patent: 4713777 (1987-12-01), Klovstad et al.
patent: 4783804 (1988-11-01), Juang et al.
patent: 5805832 (1998-09-01), Brown et al.
patent: 6029124 (2000-02-01), Gillick et al.
patent: 6839669 (2005-01-01), Gould et al.
Djuric et al.; Estimation of Nonstationary Hidden Markov Models by MCMC Sampling; Acoustics, Speech and Signal Processing, 1999, ICASSP '99, Proceedings., 1 Internationa Conference on, vol. 3, pp. 1737-1740, Mar. 1999.
“Explorations with Fabricated Data”, by Don McAllaster, Larry Gillick, Francesco Scattone, Mike Newman, fromBroadcast News Transcription Understanding Workshop, Lansdowne Conference Resort, Lansdowne, Virginia, Feb. 8-11, 1998, 4 pages.
“Why is conversational speech so hard to recognize?” by Don McAllaster, Francesco Scattone, Mike Newman, Larry Gillick, as submitted for publication toSpeech Communication, Nov. 4, 1999, 29 pages.
“Fabricating conversational Speech Data With Acoustic Models; A Program To Examine Model-Data Mismatch”, by Don McAllaster, Larry Gillick, Francesco Scattone, Mike Newman, from the Proceedings of 5th International Conference on Spoken Language Procesing, Nov. 30-Dec. 4, 1998, Sydney, Australia, pp. 1847-1850.
“Studies in Acoustic Training and Language Modeling Using Simulated Speech Data”, by Don McAllaster and Larry Gillick, from Proceedings of Eurospeech '99, 6thEuropean Conference on Speech Communication and Technology, Sep. 5-9, 1999, Budapest, Hungary, pp. 1787-1790.
“Comparative Studies of Conversational vs. Read Speech Using Simulated Acoustic Data”, by Francesco Scattone, Don McAllaster, and Larry Gillick, from th proceedings of Automatic Speech Recognition and Understanding Workshop, Keystone, Colorado, Dec. 12-15, 1999 (ASRU-99), pp. 103-106.

Affiliated with

Also associated with

Rating

Training speech recognition word models from word samples... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Training speech recognition word models from word samples..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Training speech recognition word models from word samples... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3623709