Accurately Capturing Speech Feature Distributions by Extending Supervectors for Robust Speaker Recognition.
Kevin WilkinghoffPublished in: ITG Symposium on Speech Communication (2018)
Keyphrases
- speaker recognition
- feature distributions
- gaussian mixture model
- speaker identification
- vector quantization
- speaker verification
- feature descriptors
- speech recognition
- noisy environments
- probabilistic neural network
- speech signal
- feature set
- maximum likelihood
- gray level
- neural network
- principal component analysis
- nearest neighbor