State of the art discriminative training of subspace constrained Gaussian mixture models in big training corpora.
Jing HuangPeder A. OlsenVaibhava GoelPublished in: ICASSP (2013)
Keyphrases
- gaussian mixture model
- discriminative training
- maximum likelihood
- mixture model
- em algorithm
- hidden markov models
- feature space
- expectation maximization
- posterior probability
- probabilistic interpretation
- probability density function
- feature vectors
- low dimensional
- gaussian mixture
- automatic speech recognition
- training corpus
- support vector machine
- high dimensional
- co training
- feature extraction
- maximum a posteriori
- generative model
- machine learning
- information retrieval