Maximum likelihood Linear Dimension Reduction of heteroscedastic feature for robust Speaker Recognition.
Suwon ShonSeongkyu MunDavid K. HanHanseok KoPublished in: AVSS (2015)
Keyphrases
- speaker recognition
- maximum likelihood
- gaussian mixture model
- gaussian distribution
- speaker verification
- feature vectors
- vector quantization
- speaker identification
- expectation maximization
- probabilistic neural network
- pattern recognition
- support vector machine
- image features
- em algorithm
- gaussian process regression