Prior knowledge guided MEL based model selection and adaptation for nonnative speech recognition.
Xiaodong HeYunxin ZhaoPublished in: ICASSP (1) (2004)
Keyphrases
- model selection
- speech recognition
- prior knowledge
- cepstral coefficients
- speaker adaptation
- cross validation
- hidden markov models
- language model
- speech synthesis
- speech recognizer
- pattern recognition
- speech signal
- speaker identification
- model selection criteria
- feature selection
- automatic speech recognition
- variable selection
- mixture model
- bayesian framework
- prior information
- selection criterion
- noisy environments
- bayesian information criterion
- information criterion
- speech recognition technology
- marginal likelihood
- gaussian process
- unsupervised learning
- machine learning
- speaker recognition
- generative model
- support vector
- speaker independent
- face recognition
- image segmentation