Combining the glottal mixture model (GLOMM) with UBM for speaker recognition.
Paul M. BaggenstossPublished in: EUSIPCO (2016)
Keyphrases
- mixture model
- speaker recognition
- gaussian mixture model
- speech signal
- em algorithm
- probabilistic model
- expectation maximization
- speaker identification
- generative model
- model selection
- language model
- unsupervised learning
- probabilistic neural network
- vector quantization
- speech recognition
- maximum likelihood
- gaussian processes
- data sets
- bayesian information criterion