Applications of Dirichlet Process Mixtures to speaker adaptation.
Amir Hossein Harati Nejad TorbatiJoe PiconeMarc SobelPublished in: ICASSP (2012)
Keyphrases
- dirichlet process
- speaker adaptation
- mixture model
- dirichlet processes
- maximum likelihood
- speech recognition
- em algorithm
- gaussian processes
- gaussian mixture model
- expectation maximization
- multi task learning
- language model
- automatic speech recognition
- variational inference
- sampling algorithm
- bayesian model
- posterior distribution
- probabilistic model
- generative model
- speech recognizer
- machine learning
- pattern recognition
- unsupervised learning
- dirichlet distribution
- parameter estimation
- model selection
- markov chain monte carlo
- variational methods
- latent topics
- maximum a posteriori
- bayesian framework
- closed form
- support vector
- image segmentation