Novel cascaded Gaussian mixture model-deep neural network classifier for speaker identification in emotional talking environments.
Ismail ShahinAli Bou NassifShibani HamsaPublished in: Neural Comput. Appl. (2020)
Keyphrases
- speaker identification
- gaussian mixture model
- probabilistic neural network
- neural network
- speaker recognition
- mixture model
- feature space
- feature vectors
- feature transformation
- em algorithm
- maximum likelihood
- speech recognition
- feature extraction
- learning algorithm
- face detection
- expectation maximization
- unsupervised learning
- prior knowledge
- decision trees
- feature selection