A DISCRIMATIVE TRAINING ALGORITHM FOR GAUSSIAN MIXTURE SPEAKER MODELS

5th European Conference on Speech Communication and Technology, EUROSPEECH 1997

Volumn , Issue , 1997, Pages 959-962

  He, Jialong ^a   Liu, Li ^a   Palm, Günther ^a  

^a UNIVERSITY OF ULM   (Germany)

Author keywords

[No Author keywords available]

Indexed keywords

IMAGE SEGMENTATION; SPEECH COMMUNICATION; SPEECH RECOGNITION;

DISCRIMINATIVE POWER; DISCRIMINATIVE TRAINING; EXPECTATIONS MAXIMIZATION ALGORITHMS; GAUSSIAN-MIXTURES; LIKELIHOOD CRITERION; OBSERVATION DATA; SPEAKER MODEL; TRAINING ALGORITHMS; TRAINING PROCEDURES; VECTOR GROUPS;

MAXIMUM PRINCIPLE;

EID: 0009605874     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (4)

1. D. A. Reynolds and R. C. Rose, "Robust text-independent speaker identification using Gaussian mixture speaker models,", IEEE Trans. on speech and audio processing, Vol. 2, pp. 72-83, 1995.
2. H. Z. Li, J.-P. Haton, and Y. Gong, "On MMI learning of Gaussian mixture for speaker models," Proc. of European Conference on Speech Technology (EUROSPEECH), Vol. 1, pp. 363- 366, 1995, Madrid, Spain.
3. J. He, L. Liu and G. Palm, "A new codebook training algorithm for VQ-based speaker recognition," Proc. IEEE, ICASSP'97, Vol. 2, pp. 1091-1094, Munich, GERMANY.
4. J. Godfrey, D. Graff and A. Martin, "Public databases for speaker recognition and verification," ESCA Workshop on Automatic Speaker Recognition, Identification and Verification, pp. 39-42, 1994, Martigny, Switzerland.