Fewer-token Neural Speech Codec with Time-invariant Codes.
Yong RenTao WangJiangyan YiLe XuJianhua TaoChuyuan ZhangJunzuo ZhouPublished in: CoRR (2023)
Keyphrases
- network architecture
- speech recognition
- neural network
- speech synthesis
- speech signal
- neural model
- text to speech
- video codec
- error correction
- audio visual
- bio inspired
- automatic speech recognition
- recognition engine
- bitstream
- video coding
- motion estimation
- endpoint detection
- coding method
- noisy environments
- broadcast news
- speaker identification
- dialogue system
- iris code
- error control
- error correcting codes
- decoding algorithm
- biologically inspired
- hidden markov models