Data-Driven UBM Generation via Tied Gaussians for GMM-Supervector Based Accent Identification.
Rong ZhengCe ZhangBo XuPublished in: INTERSPEECH (2011)
Keyphrases
- gaussian mixture model
- data driven
- mixture model
- gaussian mixture
- expectation maximization
- gaussian mixture modeling
- speaker recognition
- em algorithm
- maximum likelihood
- feature vectors
- speaker identification
- probability density function
- feature space
- speech recognition
- density estimation
- machine learning
- generation process
- background subtraction
- mixture of gaussians
- image retrieval
- data sets
- probabilistic model
- automatic identification
- spoken language
- loss function
- gaussian model
- decision trees
- computer vision
- neural network