HMM training based on CV-EM and CV Gaussian mixture optimization.
Takahiro ShinozakiTatsuya KawaharaPublished in: ASRU (2007)
Keyphrases
- gaussian mixture
- expectation maximization
- em algorithm
- gaussian mixture model
- hidden markov models
- mixtures of gaussians
- mixture model
- closed form
- baum welch
- mixture distributions
- probability density function
- maximum likelihood
- covariance matrix
- gaussian density
- gaussian densities
- mixture of gaussians
- probability density
- probabilistic model
- density estimation
- image reconstruction
- parameter estimation
- maximum likelihood estimation
- unsupervised learning
- upper bound
- prior knowledge
- training set
- training data