Discriminative training of Gaussian mixture models for large vocabulary speech recognition systems.
Lalit R. BahlMukund PadmanabhanDavid NahamooP. S. GopalakrishnanPublished in: ICASSP (1996)
Keyphrases
- speech recognition systems
- discriminative training
- gaussian mixture model
- speech recognition
- maximum likelihood
- hidden markov models
- automatic speech recognition
- mel frequency cepstral coefficients
- mixture model
- speech recognizer
- em algorithm
- speaker recognition
- speaker identification
- probabilistic interpretation
- expectation maximization
- noisy environments
- language model
- probability density function
- speech signal
- feature vectors
- feature space
- posterior probability
- speaker independent
- co training
- semi supervised learning
- support vector machine
- pairwise
- pattern recognition