Discriminative training of GMM for speaker identification.
Cesar Martín del AlamoFrancisco Javier Caminero GilCelinda de la Torre-MunillaLuis A. Hernández GómezPublished in: ICASSP (1996)
Keyphrases
- speaker identification
- discriminative training
- gaussian mixture model
- automatic speech recognition
- speech recognition
- hidden markov models
- maximum likelihood
- speech signal
- broadcast news
- speaker recognition
- noisy environments
- probabilistic interpretation
- language identification
- feature extraction
- posterior probability
- support vector machine
- co training
- mixture model
- feature vectors
- em algorithm
- mel frequency cepstral coefficients
- text classification
- speaker independent
- neural network
- audio features
- information extraction
- low level
- active learning