Automatic confidence score mapping for adapted speech recognition systems

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

Volumn 1, Issue , 2002, Pages

  Sankar, Ananth ^a   Kannan, Ashvin ^a  

^a NUANCE COMMUNICATIONS   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

AUTOMATIC CONFIDENCE SCORE MAPPING; AUTOMATIC SPEECH RECOGNITION; IN-GRAMMAR ERROR; MAXIMUM-A-POSTERIORI ADAPTATION; OUT-OF-GRAMMAR ERROR;

ALGORITHMS; CONFORMAL MAPPING; ERROR ANALYSIS; GRAPH THEORY; MARKOV PROCESSES; MATHEMATICAL MODELS; MAXIMUM LIKELIHOOD ESTIMATION; PATTERN RECOGNITION SYSTEMS; REGRESSION ANALYSIS;

CONTINUOUS SPEECH RECOGNITION;

EID: 0036295826     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (8)

1. Maximum likelihood linear regression for speaker adaptation of continuous density hidden markov models
2. Speaker adaptation using constrained reestimation of Gaussian mixtures
3. A maximum-likelihood approach to stochastic matching for robust speech recognition
4. Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chains
5. A phone-dependent confidence measure for utterance rejection
6. Improving rejection with semantic slot-based confidence scores
7. Task-specific adaptation of speech recognition models
8. Genones: Generalized mixture tying in continuous hidden Markov model-based speech recognizers