Adaptation experiments on the spine database with the extended maximum likelihood linear transformation (EMLLT) model

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

Volumn 1, Issue , 2002, Pages

  Gopinath, Ramesh ^a   Goel, Vaibhava ^a   Visweswariah, Karthik ^a   Olsen, Peder ^a  

^a IBM T J WATSON RESEARCH CENTER   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ALGORITHMS; DATABASE SYSTEMS; MATRIX ALGEBRA; PROBABILITY; SPEECH RECOGNITION; STATISTICS;

EXTENDED MAXIMUM LIKELIHOOD LINEAR TRANSFORMATION; SPEAKER ADAPTIVE TRAINING;

MAXIMUM LIKELIHOOD ESTIMATION;

EID: 0036298776     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/icassp.2002.5743891     Document Type: Conference Paper

References (8)

1. Extended mllt for Gaussian mixture models
2. Modeling inverse covariance matrices by basis expansion
3. Maximum likelihood linear transformations for hmm-based speech recognition
4. A compact model for speaker adaptive training
5. An introduction to the conjugate gradient method without the agonizing pain
6. Robust speech recognition in noisy environments: The ibm spine-2 evaluation system
7. Finding consensus in speech recognition: Word error minimization and other applications of confusion networks
8. Model selection in acoustic modeling