Combined time-domain harmonic compression and CVSD for 7.2 kbit/s transmission of speech signals.
David MalahPublished in: ICASSP (1980)
Keyphrases
- speech signal
- speech recognition
- automatic speech recognition
- automatic speech recognition systems
- video signals
- speaker dependent
- speaker identification
- noisy environments
- vocal tract
- compression ratio
- compression algorithm
- non stationary
- background noise
- hidden markov models
- linear prediction
- image compression
- neural network
- spectral analysis
- sound source
- speech enhancement
- blind source separation
- speaker recognition
- acoustic features
- multiresolution
- multiscale