Very low bit rate speech coding based on HMM with speaker adaptation.
Takashi MasukoTakao KobayashiKeiichi TokudaPublished in: Systems and Computers in Japan (2006)
Keyphrases
- speaker adaptation
- speech recognition
- automatic speech recognition
- maximum likelihood
- speaker independent
- hidden markov models
- speech signal
- speech recognizer
- coding scheme
- video coding
- speaker dependent
- speech synthesis
- video coder
- vocal tract
- speaker identification
- linear prediction
- broadcast news
- noisy environments
- language model
- bit rate
- information retrieval
- spoken language
- speaker recognition
- multi stream
- neural network
- inter frame