Eigen-channel compensation and discriminatively trained Gaussian mixture models for dialect and accent recognition.
Pedro A. Torres-CarrasquilloDouglas E. SturimDouglas A. ReynoldsAlan McCreePublished in: INTERSPEECH (2008)
Keyphrases
- gaussian mixture model
- discriminatively trained
- mixture model
- speaker recognition
- object detection
- em algorithm
- feature vectors
- object recognition
- maximum likelihood
- pattern recognition
- feature space
- speaker identification
- expectation maximization
- discriminative training
- generative model
- feature extraction
- discriminative learning
- probability density function
- speech recognition
- unsupervised learning
- probabilistic model
- image matching
- learning algorithm
- principal component analysis
- bayesian networks
- k means