Patent
1992-12-11
1995-08-08
MacDonald, Allen R.
395 265, G10L 918
Patent
active
054406625
ABSTRACT:
A two-pass classification system and method that post-processes HMM scores with additional confidence scores to derive a value that may be applied to a threshold on which a keyword verses non-keyword determination may be based. The first stage comprises Generalized Probabilistic Descent (GPD) analysis which uses feature vectors of the spoken words and the HMM segmentation information (developed by the HMM detector during processing) as inputs to develop a first set of confidence scores through a linear combination (a weighted sum) of the feature vectors of the speech. The second stage comprises a linear discrimination method that combines the HMM scores and the confidence scores from the GPD stage with a weighted sum to derive a second confidence score. The output of the second stage may then be compared to a predetermined threshold to determine whether the spoken word or words include a keyword.
REFERENCES:
patent: 4713777 (1987-12-01), Kloustad et al.
patent: 5125022 (1992-06-01), Hunt et al.
patent: 5127043 (1992-06-01), Hunt et al.
patent: 5199077 (1993-03-01), Wilcox et al.
patent: 5303299 (1994-04-01), Hunt et al.
"Accessing Custom Calling Telephone Services Using Speech Recognition", by R. A. Sukkar & K. V. Kinder, pp. 994-999.
"Speech Recognition Using Segmented Neural Nets", by S. Austin, G. Zavaliagkos, J. Makhoul, R. Schwartz, IEEE 0-7803-0532-9/92, 1992, pp. I-625-I-628.
"Continuous Word Spotting For Applications In Telecommunications", by M-W Feng & B. Mazor, Tu.sAM.1.5, pp. 21-24.
"Discriminative Analysis For Feature Reduction In Automatic Speech Recognition", by E. L. Bocchieri & J. G. Wilpon, 0-7803-0532-9/92, IEEE 1992, pp. I-501-I-504.
"Segmental GPD Training Of HMM Based Speech Recognizer", by W. Chow, B. H. Juang, C. H. Lee, 0-7803-0532-9/92, IEEE 1992, pp. I-473-I-476.
"Improvements in Connected Digit Recognition Using Higher Order Spectral and Energy Features", by J. G. Wilpon, C-H. Lee, L. R. Rabiner, CH2977-7/91/0000-0349, 1991 IEEE, pp. 349-352.
"Automatic Recognition of Keywords in Unconstrained Speech Using Hidden Markov Models", by J. G. Wilpon, L. R. Rabiner, C-H. Lee, E. R. Goldman, 0096-3518/90/1100-1870$01.00, 1990 IEEE, pp. 1870-1878.
"A Hidden Markov Model Based Keyword Recognition System", by R. C. Rose & D. B. Paul, CH2847-2/90/0000-0129, 1990 IEEE, pp. 129-131.
"An Introduction to Hidden Markov Models", by L. R. Rabiner & B. H. Juang, IEEE ASSP Magazine, Jan. 1986, pp. 4-16.
AT&T Corp.
Johannesen Michael B.
MacDonald Allen R.
Onka T.
LandOfFree
Keyword/non-keyword classification in isolated word speech recog does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Keyword/non-keyword classification in isolated word speech recog, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Keyword/non-keyword classification in isolated word speech recog will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-977583