Comparative experiments of several adaptation approaches to noisy speech recognition using stochastic trajectory models

Speech Communication

Volumn 18, Issue 4, 1996, Pages 335-352

  Siohan, Olivier ^a,b   Gong, Yifan ^a   Haton, Jean Paul ^a  

^a CNRS   (France)

^b LUCENT TECHNOLOGIES   (United States)

Author keywords

Adaptation by linear regression; Model combination; Noisy speech recognition; Space mapping; State based filtering; Stochastic trajectory models

Indexed keywords

ACOUSTIC NOISE; ACOUSTIC SURFACE WAVE FILTERS; COMPUTER SIMULATION; REGRESSION ANALYSIS; SIGNAL TO NOISE RATIO; SPEECH ANALYSIS; SPURIOUS SIGNAL NOISE;

CONTEXT INDEPENDENT PHONE MODELS; LINEAR REGRESSION METHODS; STOCHASTIC TRAJECTORY MODELS (STM); VOCABULARY INDEPENDENT ACOUSTIC TRAINING; WORD ERROR RATE; WORD PAIR PERPLEXITY;

SPEECH RECOGNITION;

EID: 0030165492     PISSN: 01676393     EISSN: None     Source Type: Journal    
DOI: 10.1016/0167-6393(96)00015-5     Document Type: Article

References (19)

1. A.P. Dempster, N.M. Laird and D.B. Rubin (1977), "Maximum likelihood from incomplete data via the EM algorithm", J. Roy. Statist. Soc. Ser. B, Vol. 39, pp. 1-39.
2. Y. Ephraim and D. Malah (1985), "Speech enhancement using a minimum mean-square error log-spectral amplitude estimator", IEEE Trans. Acoust. Speech Signal Process., Vol. 33, No. 2, pp. 443-445.
3. A. Erell and M. Weintraub (1993), "Energy condition spectral estimation for recognition of noisy speech", IEEE Trans. Speech Audio Process., Vol. 1, No. 1, pp. 84-89.
4. M.J.F. Gales and S.J. Young (1993), "Cepstral parameter compensation for HMM recognition in noise", Speech Communication, Vol. 12, No. 3, pp. 231-239.
5. Y. Gong (1995), "Speech recognition in noisy environments: A survey", Speech Communication, Vol. 16, No. 3, pp. 261-291.
6. Y. Gong and J.-P. Haton (1992), "DTW-based phonetic labeling using explicit phoneme duration constraints", Proc. Internat. Conf. on Spoken Language Processing, Banff, Alberta, Canada, ICSLP'92, Vol. 2, pp. 863-866.
7. Y. Gong and J.-P. Haton (1993), "Iterative transformation and alignment for speech labeling", Proc. European Conf. on Speech Communication and Technology, Berlin, Germany, EUROSPEECH'93, Vol. 3, pp. 1759-1762.
8. Y. Gong and J.-P. Haton (1994), "Stochastic trajectory modeling for speech recognition", Proc. IEEE Internat. Conf. Acoust. Speech Signal Process., Adelaide, Australia, ICASSP'94, Vol. 1, pp. 57-60.
9. R.M. Gray, A. Buzo, A.H. Gray and Y. Matsuyama (1980), "Distorsion measures for speech processing", IEEE Trans. Acoust. Speech Signal Process., Vol. 28, pp. 367-376.
10. C.J. Leggetter and P.C. Woodland (1994), Speaker adaptation of HMMs using linear regression. Technical Report F-INFENG/TR. 181, CUED, Cambridge University, Engineering Department, UK.
11. C. Montacié and G. Chollet (1988), "Evaluating speech recognizers and databases", in Recent Advances in Speech Understanding and Dialog Systems, ed. by H. Niemann, NATO ASI Series, Vol. F46.
12. O. Siohan, Y. Gong and J.-P. Haton (1994), "A comparison of three noisy speech recognition approaches", Proc. Internat. Conf. on Spoken Language Processing, Yokohama, Japan, ICSLP'94, Vol. 3, pp. 1031-1034.
13. O. Siohan, Y. Gong and J.-P. Haton (1995), "Noise adaptation using linear regression for continuous noisy speech recognition", Proc. of European Conf. on Speech Communication and Technology, Madrid, Spain, EUROSPEECH'95.
14. W.C. Treurniet and Y. Gong (1994), "Noise independent speech recognition for a variety of noise types", Proc. IEEE Internat. Conf. Acoust. Speech Signal Process., Adelaide, Australia, ICASSP'94, Vol. 1, pp. 437-440.
15. D. Van Compernolle (1989), "Spectral estimation using a log-distance error criterion applied to speech recognition", Proc. IEEE Internat. Conf. Acoust. Speech Signal Process., Glasgow, UK, ICASSP'89, pp. 258-261.
16. A.P. Varga, H.J.M. Steeneken, M. Tomlinson and D. Jones (1992), The NOISEX-92 study on the effect of additive noise on automatic speech recognition, Technical report. DRA Speech Research Unit.
17. F. Xie and D. Van Compernolle (1993), "Speech enhancement by nonlinear spectral estimation - A unifying approach", Proc. European Conf. on Speech Communication and Technology, Berlin, Germany, EUROSPEECH'93, Vol. 1, pp. 617-620.
18. F. Xie and D. Van Compernolle (1994), "A family of MLP based nonlinear spectral estimators for noise reduction", Proc. IEEE Internat. Conf. Acoust. Speech Signal Process., Adelaide, Australia, ICASSP'94, Vol. 2, pp. 53-56.
19. S.J. Young (1992), HTK: Hidden Markov model toolkit v1.4 reference manual, Technical report, Speech Group, Cambridge University, Engineering Department.