Speaker Clustering for a Mixture of Singing and Reading.
Mahnoosh MehrabaniJohn H. L. HansenPublished in: INTERSPEECH (2012)
Keyphrases
- speaker clustering
- gaussian mixture model
- mixture model
- discriminant analysis
- partially supervised
- dimensionality reduction
- expectation maximization
- em algorithm
- feature transformation
- feature vectors
- cosine distance
- feature space
- model selection
- probabilistic model
- high dimensional
- multiscale
- maximum likelihood
- feature extraction
- generative model
- unsupervised learning
- distance measure
- high dimensional data
- multi class
- linear classifiers
- multi instance
- acoustic features
- pairwise
- similarity measure