Application of variational Bayesian PCA for speech feature extraction

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

Volumn 1, Issue , 2002, Pages

  Kwon, Oh Wook ^a   Lee, Te Won ^a   Chan, Kwokleung ^a  

^a UNIVERSITY OF CALIFORNIA SAN DIEGO   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

BAYESIAN PRINCIPAL COMPONENT ANALYSIS METHOD; GAUSSIAN MIXTURES; MEL-FREQUENCY CEPSTRAL COEFFICIENT;

ACOUSTIC SIGNAL PROCESSING; FEATURE EXTRACTION; MATRIX ALGEBRA; MAXIMUM LIKELIHOOD ESTIMATION; PRINCIPAL COMPONENT ANALYSIS; VARIATIONAL TECHNIQUES;

SPEECH RECOGNITION;

EID: 0036294874     PISSN: 15206149     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (11)

1. A study of two dimensional linear discriminants for ASR
2. Peripheral features for HMM-based speech recognition
3. Integration of fixed and multiple resolution analysis in a speech recognition system
4. Probabilistic principal component analysis
5. Mixtures of probabilistic principal component analysers
6. A unifying review of linear Gaussian models
7. Variational principal components
8. Bayesian PCA
9. Variational inference for Bayesian mixtures of factor analysers
10. Variational learning of clusters of undercomplete nonsymmetric independent components