A switching linear Gaussian hidden Markov model and its application to nonstationary noise compensation for robust speech recognition

EUROSPEECH 2003 - 8th European Conference on Speech Communication and Technology

Volumn , Issue , 2003, Pages 977-980

  Wu, Jian ^a   Huo, Qiang ^a  

^a UNIVERSITY OF HONG KONG   (Hong Kong)

Author keywords

[No Author keywords available]

Indexed keywords

BAYESIAN NETWORKS; GAUSSIAN DISTRIBUTION; HIDDEN MARKOV MODELS; MARKOV PROCESSES; SPEECH; SPEECH COMMUNICATION; TRELLIS CODES;

LINEAR GAUSSIAN MODEL; LINGUISTIC INFORMATION; MULTI-CONDITION TRAININGS; NOISY SPEECH RECOGNITION; NONSTATIONARY DISTORTIONS; ROBUST SPEECH RECOGNITION; SWITCHING CONDITIONS; SWITCHING LINEAR GAUSSIAN HIDDEN MARKOV MODELS;

SPEECH RECOGNITION;

EID: 85009212133     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (8)

1. M.J.F. Gales, "Adaptive training for robust ASR," Proc. IEEE ASRU-01, Italy, 2001.
2. Z. Ghahramani and G.E. Hinton, "Variational learning for switching state-space models," Neural Computation, vol. 12, no. 4, pp. 831-864, 2000.
3. H. G. Hirsch and D. Pearce, "The AURORA experimental framework for the performance evaluations of speech recognition systems under noisy conditions", ISCA ITRW ASR2000, Paris, France, September 2000.
4. Q. Huo and B.Ma, "Robust speech recognition based on offline elicitation of multiple priors and on-line adaptive prior fusion," Proc. ICSLP-2000, Beijing, China, 2000.
5. F.V. Jensen, S.L. Lauritzen and K.G. Olesen, "Bayesian updating in recursive graphical models by local computations," Computational Statistical Quarterly, vol. 4, pp. 269-282, 1990.
6. S. Roweis and Z. Ghahramani, "A unifying review of Linear Gaussian Models," Neural Computation, vol. 11, pp. 305- 345, 1999
7. J. Wu and Q. Huo, "An environment compensated minimum classification error training approach and its evaluation on Aurora2 database," Proc. ICSLP-2002, 2002, pp.I-453-456.
8. J. Wu and Q. Huo, "Modelling uncertainty in stochastic vector mapping with minimum classification error training for robust speech recognition," Proc. ICASSP-2003, HK, 2003.