Variable dimension vector quantization of linear predictive coefficients of speech.
Philip A. ChouTom D. LookabaughPublished in: ICASSP (1) (1994)
Keyphrases
- vector quantization
- linear predictive
- cepstral coefficients
- speech recognition
- speaker recognition
- mel frequency cepstral coefficients
- speech signal
- image compression
- linear prediction
- feature set
- vector quantizer
- automatic speech recognition
- hidden markov models
- audio signal
- language model
- speaker identification
- acoustic features
- maximum likelihood
- gaussian mixture model
- neural network
- subband
- non stationary
- pattern recognition