Initializing Subspace Constrained Gaussian Mixture Models.
Peder A. OlsenKarthik VisweswariahRamesh GopinathPublished in: ICASSP (1) (2005)
Keyphrases
- gaussian mixture model
- feature space
- covariance matrices
- mixture model
- feature vectors
- maximum likelihood
- speaker recognition
- gaussian mixture
- em algorithm
- dimensionality reduction
- speaker identification
- low dimensional
- probability density
- probability density function
- variational bayes
- expectation maximization
- feature extraction
- model selection
- density estimation
- high dimensional data
- principal component analysis
- high dimensional
- covariance matrix
- unsupervised learning
- gaussian distribution
- learning algorithm
- data points
- bayesian networks
- image processing