ROBUST SPEECH RECOGNITION USING HMM'S WITH TOEPLITZ STATE COVARIANCE MATRICES

5th International Conference on Spoken Language Processing, ICSLP 1998

Volumn , Issue , 1998, Pages

  Roberts, William J J ^a   Ephraim, Yariv ^b  

^a DEFENCE SCIENCE AND TECHNOLOGY ORGANISATION   (Australia)

^b George Mason University   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

COVARIANCE MATRIX; HIDDEN MARKOV MODELS; MAXIMUM PRINCIPLE; SIGNAL TO NOISE RATIO; SPEECH RECOGNITION;

GAUSSIANS; NOISY SIGNALS; ROBUST SPEECH RECOGNITION; SHORT SIGNAL; SIGNAL VECTORS; SPEECH SIGNALS; SPEECH WAVEFORMS; STATE-COVARIANCE MATRIX; TOEPLITZ; TOEPLITZ COVARIANCE MATRIXES;

MAXIMUM LIKELIHOOD ESTIMATION;

EID: 0011450087     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (5)

1. Y. Ephraiin, '"Gain-adapted hidden Markov models for recognition of clean and noisy speech,'' IEEE Trans, on Speech Proc, vol. 40, no. 6, pp. 1303-1316, June 1992.
2. A. Dembo, C. L. Mallows, and L. A. Shepp, "Embedding non-negative definite Toeplitz matrices in non-negative definite circulant matrices, with application to covariance estimation,'*' IEEE Trans, on Inform.. Theory, vol. 35, no. 6, pp. 1206-1212, Dec. 1989.
3. M. I. Miller and D. L. Snyder, "The role of likelihood and entropy in incomplete-data problems. Applications to estimating point-process intensities and Toeplitz constrained covariances." Proc. of the IEEE. vol. 75, no. 7, pp. 892-907, July 1987.
4. 1' submitted for 'publication.
5. M. J. F. Gales and S. J. Young, "Robust continuous speech recognition using parallel model combination," IEEE Trans, on Speech and Audio Processing, vol. 4, no. 5, pp. 352-359, Sept. 1996.