Enhancing quality of CELP coded speech via wideband extension by using voicing GMM interpolation and HNM re-synthesis.
Dar Ghulam RazaCheung-Fat ChanPublished in: ICASSP (2002)
Keyphrases
- speech quality
- speech signal
- linear predictive
- speaker identification
- speaker recognition
- speech recognition
- high quality
- gaussian mixture model
- packet loss
- pattern recognition
- automatic speech recognition
- hidden markov models
- signal to noise ratio
- non stationary
- voice activity detection
- linear prediction
- text to speech
- mel frequency cepstral coefficients
- video sequences
- expectation maximization