Subspace constrained Gaussian mixture models for speech recognition.
Scott AxelrodVaibhava GoelRamesh A. GopinathPeder A. OlsenKarthik VisweswariahPublished in: IEEE Trans. Speech Audio Process. (2005)
Keyphrases
- gaussian mixture model
- speech recognition
- speaker identification
- speaker recognition
- feature space
- covariance matrices
- mixture model
- language model
- speech signal
- hidden markov models
- speech synthesis
- pattern recognition
- feature vectors
- maximum likelihood
- em algorithm
- speaker independent
- automatic speech recognition
- speech recognition systems
- speech recognizer
- mel frequency cepstral coefficients
- low dimensional
- principal component analysis
- bayesian information criterion
- noisy environments
- feature extraction
- speech recognition technology
- probability density function
- high dimensional
- probabilistic neural network
- audio signal
- speaker diarization
- dimensionality reduction
- neural network