Combining Gaussian Mixture Models and Segmental Feature Models for Speaker Recognition.
Milana MilosevicUlrike GlavitschPublished in: INTERSPEECH (2017)
Keyphrases
- speaker recognition
- gaussian mixture model
- vector quantization
- feature vectors
- probabilistic neural network
- mixture model
- speaker verification
- speaker identification
- maximum likelihood
- probabilistic model
- expectation maximization
- speech signal
- hidden markov models
- learning algorithm
- language model
- feature space
- feature extraction
- image processing