Electrical audio signal processing systems and devices – One-way audio signal program distribution – Public address system
Patent
1989-03-23
1990-02-20
Harkcom, Gary V.
Electrical audio signal processing systems and devices
One-way audio signal program distribution
Public address system
381 43, G10L 906
Patent
active
049033050
ABSTRACT:
A method is provided for deriving acoustic word representations for use in speech recognition. Initial word models are created, each formed of a sequence of acoustic sub-models. The acoustic sub-models from a plurality of word models are clustered, so as to group acoustically similar sub-models from different words, using, for example, the Kullback-Leibler information as a metric of similarity. Then each word is represented by cluster spelling representing the clusters into which its acoustic sub-models were placed by the clustering. Speech recognition is performed by comparing sequences of frames from speech to be recognized against sequences of acoustic models associated with the clusters of the cluster spelling of individual word models. The invention also provides a method for deriving a word representation which involves receiving a first set of frame sequences for a word, using dynamic programming to derive a corresponding initial sequence of probabilistic acoustic sub-models for the word independently of any previously derived acoustic model particular to the word, using dynamic programming to time align each of a second set of frame sequences for the word into a succession of new sub-sequences corresponding to the initial sequence of models, and using these new sub-sequences to calculate new probabilistic sub-models.
REFERENCES:
patent: 4590605 (1986-05-01), Hataoka et al.
Wilpon et al, "A Modified K-Means Clustering Algorithm for Use in Isolated Word Recognition", IEEE Trans. on ASSP, vol. ASSP-33, No. 3, Jun. 1985.
Jelinek, "Continuous Speech Recognition by Statistical Methods", Proc. of IEEE, vol. 64, No. 4, Apr. 1976.
Bourlard et al, "Speaker Dependent Connected Speech Recognition Via Phonemic Markov Models", ICASSP 85 IEEE, vol. 3 of 4, pp. 1213-1216, Mar. 1985.
James K. Baker, "Stochastic Modeling for Automatic Speech Understanding", an article from Speech Recognition, edited by D. R. Reddy and published by Academic Press, N.Y.C., in 1972.
Janet M. Baker, "Automatic Prototype Selection for Continuous Speech Recognition", an article published in the collection of papers presented at the 97th Meeting of the Accoustical Society of America.
Janet M. Baker, "Performance Statistics of the Hear Acoustic Processor", 1979 IEEE Int. Conf. on Acoustics, Speech & Signal Processing, 79CH1379-7 ASSP, p. 262.
Burton et al., "Isolated-Word Speech Recognition Using Multisection Vector Quantization Codebooks", IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. ASSP-33, No. 4, Aug. '85, p. 837.
Kopec et al., "Network-Based Isolated Digit Recognition Using Vector Quantization", IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. ASSP-33, No. 4, Aug. '85, p. 850.
Baker James K.
Baker Janet M.
Gillick Laurence
Roth Robert S.
Sturtevant Dean
Dragon Systems, Inc.
Harkcom Gary V.
Knepper David D.
Porter Edward W.
LandOfFree
Method for representing word models for use in speech recognitio does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for representing word models for use in speech recognitio, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for representing word models for use in speech recognitio will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1622486