Continuous mixture densities and linear discriminant analysis for improved context-dependent acoustic models.
Xavier L. AubertReinhold Haeb-UmbachHermann NeyPublished in: ICASSP (2) (1993)
Keyphrases
- linear discriminant analysis
- context dependent
- gaussian densities
- discriminant analysis
- face recognition
- dimensionality reduction
- principal component analysis
- feature space
- feature extraction
- support vector
- high dimensional data
- support vector machine svm
- mixture model
- null space
- natural language
- high level
- subspace methods
- machine learning
- speech recognition
- low level
- acoustic models
- data points
- decision trees
- bayesian networks
- knn
- feature selection