Power-Normalized Cepstral Coefficients (PNCC) for robust speech recognition.
Chanwoo KimRichard M. SternPublished in: ICASSP (2012)
Keyphrases
- speech recognition
- cepstral coefficients
- speaker diarization
- noisy environments
- hidden markov models
- speech signal
- automatic speech recognition
- pattern recognition
- language model
- speaker identification
- feature set
- image coding
- mel frequency cepstral coefficients
- speaker recognition
- neural network
- bayesian networks
- computer vision