Speaker indexing and adaptation using speaker clustering based on statistical model selection.
Masafumi NishidaTatsuya KawaharaPublished in: ICASSP (1) (2004)
Keyphrases
- model selection
- statistical inference
- cross validation
- covariate shift
- hyperparameters
- parameter estimation
- sample size
- speaker adaptation
- speech recognition
- statistical learning
- mixture model
- motion segmentation
- machine learning
- model selection criteria
- multivariate regression
- feature selection
- bayesian learning
- regression model
- hypothesis tests
- variable selection
- error estimation
- gaussian process
- audio visual
- leave one out cross validation
- bayesian model selection
- marginal likelihood
- information criterion
- selection criterion
- generalization error
- bayesian methods
- closed form
- hypothesis testing
- probabilistic model
- decision trees