Patent
1994-07-18
1998-10-20
MacDonald, Allen R.
395 254, G10L 900
Patent
active
058259783
ABSTRACT:
In accordance with the invention, a speech recognizer is provided which uses a computationally-feasible method for constructing a set of Hidden Markov Models (HMMs) for speech recognition that utilize a partial and optimal degree of mixture tying. With partially-tied HMMs, improved recognition accuracy of a large vocabulary word corpus as compared to systems that use fully-tied HMMs is achieved with less computational overhead than with a fully untied system. The computationally-feasible technique comprises the steps of determining a cluster of HMM states that share Gaussian components which are close together, developing a subset codebook for those clusters, and recalculating the Gaussians in the codebook to best estimate the clustered states.
REFERENCES:
patent: 4587670 (1986-05-01), Levinson et al.
patent: 4741036 (1988-04-01), Bahl et la.
patent: 4783804 (1988-11-01), Juang et al.
patent: 4817156 (1989-03-01), Bahl et al.
patent: 4903305 (1990-02-01), Gillick et al.
patent: 5075896 (1991-12-01), Wilcox et al.
patent: 5172228 (1992-12-01), Israelsen
patent: 5193142 (1993-03-01), Zhao
Digalakis et al., "Acoustic Calibration and Search in SRI's Large Vocabulary Recognition System," Proc. IEEE ASR Workshop, Snowbird, Dec. 1993.
Manikopoulos, "Finite State Vector Quantisation with Neural Network Classification of States," IEEE Proceedings-F, vol. 140, No. 3, Jun. 1993.
Knutson et al., "Feature Based Compression of Vector Quantized Codebooks and Data for Optimal Image Compression," Circuits and Systems, May 1993 IEEE International Symposium, May 1993.
Young, "The General Use of Tying in Phoneme-Based HMM Speech Recognisers," Proc. ICASSP, pp. I-569-I-572, Mar. 1992.
Hwang et al., "Subphonetic Modeling with Markov States--Senone," Proc. ICASSP, pp. I-33-I-36, Mar. 1992.
Lee et al., "Allophone Clustering For Continuous Speech Recognition," ICASSP '90: Acoustics, Speech & Signal Processing Conference, Feb. 1990.
Lee, "Context-Dependent Phonetic HMM For Speaker-Independent Continous Speech Recognition," IEEE Trans. ASSP, pp. 599-609, Apr. 1990.
Kubrick et al., "Classified Vector Quantisation of Images: Codebook Design Algorithm," IEE Proceedings, vol. 137, Pt. I, No. 6, Dec. 1990.
Rabiner, "A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition," Proc. IEEE, vol. 77, No. 2, pp. 267-296, Feb. 1989.
Gray, "Vector Quantization," The ASSP Magazine, vol. 1, No. 2, pp. 3-29 (Apr. 1984).
L. R. Rabiner, B. H. Juang, S. E. Levinson, and M. M. Sondhi, "Recognition of Isolated Digits Using Hidden Markov Models with Continuous Mixture Densities," Bell Systems Tech. Journal, vol. 64(6), pp. 1211-1234, 1985.
X. D. Huang, and M. A. Jack, "Performance Comparison Between Semi-continuous and Discrete Hidden Markov Models," IEEE Electronics Letters, vol. 24 No. 3, pp. 149-150.
J. R. Bellegarda and D. Nahmoo, "Tied Mixture Continuous Parameter Modeling for Speech Recognition," IEEE Trans. ASSP, vol. 38(12), pp. 2033-2045, Dec. 1990.
C. Lee, L. Rabiner, R. Pieraccini and J. Wilpon, "Acoustic Modeling for Large Vocabulary Speech Recognition," Computer Speech and Language, Apr. 1990, pp. 127-165.
D. Pallett, "Results for the Sep. 1992 Resource Management Benchmark," DARPA Workshop on Artificial Neural Networks and CSR, Sep. 1992.
D. B. Paul and E. A. Martin, "Speaker Stress-resistant Continuous Speech Recognition," Proc. ICASSP, pp. 283-286, Apr. 1988.
K. F. Lee, "Context-Dependent Phonetic Hidden Markov Models for Speaker-Independent Continuous Speech Recognition," IEEE Trans. ASSP, pp. 599-609, Apr. 1990.
L. R. Bahl, P. V. de Souza, P. S. Gopalakrishnan, D. Naha- moo and M. A. Picheny, "Context Dependent Modeling of Phones in Continuous Speech Using Decision Trees," DARPA Workshop on Speech and Natural Language, pp. 264-269, Feb. 1991.
M.-Y. Hwang and X. D. Huang, "Subphonetic Modeling with Markov States--Senone," Proc. ICASSP, pp. I-33-36, Mar. 1992.
H. Murveit, J. Butzberger, V. Digalakis and M. Weintraub, "Large Vocabulary Dictation using SRI's Deciphertm Speech Recognition System: Progressive Search Techniques," Proc. ICASSP, pp. II-319-II-322, Apr. 1993.
S. J. Young, "The General Use of Tying in Phoneme-Based HMM Speech Recognizers," Proc. ICASSP, pp. I-569-I-572, Mar. 1992.
R. Haeb-Umbach and H. Ney, "Linear Discriminant Analysis for Improved Large Vocabulary Continuous Speech Recognition," Proc. ICASSP, pp. I-13-I-16, Mar. 1992.
J. L. Gauvain and C. H. Lee, "Bayesian Learning of Gaussian Mixture Densities for Hidden Markov Models," Proc. DARPA Speech and Natural Language Workshop, Feb. 1991.
V. Digalakis, P. Monaco and H. Murveit, "Acoustic Calibration and Search in SRI's Large Vocabulary HMM-based Speech Recognition System," Proc. IEEE ASR Workshop, Snowbird, Dec. 1993.
K. F. Lee and H. W. Hon, "Speaker Independent Phone Recognition Using Hidden Markov Models," IEEE Trans. ASSP, pp. 1641-1648, 1989.
D. Pallet, J. G. Fiscus, W. M. Fisher, and J, S. Garofolo, "Benchmark Tests for the DARPA Spoken Language Program," HLT Workshop, Princeton, Mar. 1993 .
Digalakis Vassilios
Murveit Hy
Allen, Esq. Kenneth R.
LeBlanc, Esq. Stephen J.
MacDonald Allen R.
SRI - International
Wieland Susan
LandOfFree
Method and apparatus for speech recognition using optimized part 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 and apparatus for speech recognition using optimized part, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for speech recognition using optimized part will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-254883