Improving speaker verification with Figure of Merit training

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

Volumn 1, Issue , 2002, Pages 693-696

  Li, Xiaohan ^a   Chang, Eric ^b   Dai, Bei Qian ^a  

^a UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA   (China)

^b MICROSOFT RESEARCH ASIA   (China)

Author keywords

[No Author keywords available]

Indexed keywords

EXPECTATION-MAXIMIZATION ALGORITHM; FIGURE OF MERIT TRAINING; GAUSSIAN MIXTURE MODEL; RECEIVER OPERATING CHARACTERISTIC; SPEAKER VERIFICATION;

ACOUSTIC SIGNAL PROCESSING; ALGORITHMS; ERROR DETECTION; GRAPH THEORY; MATHEMATICAL MODELS; MAXIMUM LIKELIHOOD ESTIMATION; SPEECH PROCESSING; SPEECH SYNTHESIS; TEXT PROCESSING;

SPEECH RECOGNITION;

EID: 17544399814     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2002.5743812     Document Type: Article

References (9)

1. The NIST 1999 speaker recognition evaluation - An overview
2. Speaker verification using minimum verification error training
3. Comparison of background normalization methods for text-independent speaker verification
4. Speaker verification using adapted Gaussian mixture models
5. Improving wordspotting performance with limited training data
6. High-performance low-complexity wordspotting using neural networks
7. A discriminative training algorithm for Gaussian mixture speaker models
8. Speaker identification using minimum classification error training