Glottal mixture model (GLOMM) for speaker identification on telephone channels.
Paul M. BaggenstossKevin WilkinghoffFrank KurthPublished in: EUSIPCO (2017)
Keyphrases
- speaker identification
- mixture model
- speech signal
- gaussian mixture model
- speech recognition
- language model
- em algorithm
- probabilistic model
- generative model
- expectation maximization
- automatic speech recognition
- maximum likelihood
- model selection
- unsupervised learning
- non stationary
- hidden markov models
- noisy environments
- neural network
- probabilistic mixture model
- active learning
- pattern recognition