Environmental robust speech and speaker recognition through multi-channel histogram equalization.
Stefano SquartiniEmanuele PrincipiRudy RotiliFrancesco PiazzaPublished in: Neurocomputing (2012)
Keyphrases
- speaker recognition
- multi channel
- histogram equalization
- gaussian mixture model
- vector quantization
- speaker verification
- image enhancement
- speaker identification
- probabilistic neural network
- single channel
- speech signal
- speech recognition
- contrast enhancement
- noisy environments
- gray level
- image compression
- gray scale
- multiresolution
- dynamic range
- mixture model
- feature selection