Supervised adaptation of MCE-trained CDHMMS using minimum classification error linear regression

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 1, Issue , 2002, Pages

  Wu, Jian ^a   Huo, Qiang ^a  

^a UNIVERSITY OF HONG KONG   (Hong Kong)

Author keywords

[No Author keywords available]

Indexed keywords

MARKOV PROCESSES; MATHEMATICAL MODELS; MAXIMUM LIKELIHOOD ESTIMATION; PARAMETER ESTIMATION; REGRESSION ANALYSIS;

HIDDEN MARKOV MODELS; MINIMUM CLASSIFICATION ERROR LINEAR REGRESSION; SUPERVISED SPEAKER ADAPTATION;

SPEECH RECOGNITION;

EID: 0141517899     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/icassp.2002.5743790     Document Type: Conference Paper

References (14)

1. Speaker adaptation using discriminative linear regression on time-varying mean parameters in trended HMM
2. Mean and variance adaptation within the MLLR framework
3. Maximal rank likelihood as an optimization function for speech recognition
4. Discriminative speaker adaptation with conditional maximum likelihood linear regression
5. On-line adaptive learning of continuous density hidden Markov models based on multiple-stream prior evolution and posterior pooling
6. Discriminative learning for minimum error classification
7. A combination of discriminative and maximum likelihood techniques for noise robust speech recognition
8. On adaptive decision rules and decision parameter adaptation for automatic speech recognition
9. Maximum likelihood linear regression for speaker adaptation of continuous density hidden Markov models
10. A study on speaker adaptation for Mandarin syllable recognition with minimum error discriminative training
11. Improvements in linear transform based speaker adaptation
12. Lattice-based discriminative training for large vocabulary speech recognition
13. Frame-discriminative and confidence-driven adaptation for LVCSR