A hybrid GMM/SVM approach to speaker identification

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

Volumn 1, Issue , 2001, Pages 417-420

  Fine, Shai ^a   Navrátil, Jirí ^a   Gopinath, Ramesh A ^a  

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

Author keywords

[No Author keywords available]

Indexed keywords

GAUSSIAN MIXTURE MODEL; SUPPORT VECTOR MACHINES; TEXT-INDEPENDENT SPEAKER RECOGNITION;

CALCULATIONS; FEATURE EXTRACTION; MATHEMATICAL MODELS; OPTIMIZATION; SPEECH ANALYSIS; STATISTICAL METHODS;

SPEECH RECOGNITION;

EID: 0034849711     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2001.940856     Document Type: Article

References (15)

1. A training algorithm for optimal margin classifiers
2. Multigrained data modeling for speaker recognition with sparse training and test data
3. Very large population text-independent speaker identification
4. Solving multiclass learning problems via error-correcting output codes
5. Maximum a posteriori estimation for multivariate gaussian mixture observations of markov chains
6. Classification by pairwise coupling
7. Exploiting generative models in discriminative classifiers
8. Natural regularization from generative models
9. Fast trining support vector machines using sequential mininal optimization
10. Probabilities for sv machines
11. Htimit and llhdb: Speech corpora for the study of handset transducer effects
12. The lincoln speaker recognition system: Nist eval2000
13. Speaker recognition in two-speaker data: Recent results from dragon systems