Very low bit-rate F0 coding for phonetic vocoders using MSD-HMM with quantized F0 symbols.
Takashi NoseTakao KobayashiPublished in: Speech Commun. (2012)
Keyphrases
- speech recognition
- hidden markov models
- video coding
- coding scheme
- speaker independent
- video coder
- coding method
- transform coefficients
- handwritten word recognition
- handwritten words
- pattern recognition
- hand drawn
- low bit rate
- language model
- gesture recognition
- significant bit
- acoustic models
- speaker adaptation
- hierarchical hidden markov model
- speech signal
- multi layer
- motion compensation
- bit rate
- optical flow