WNARS: WFST based Non-autoregressive Streaming End-to-End Speech Recognition.
Zhichao WangWenwen YangPan ZhouWei ChenPublished in: CoRR (2021)
Keyphrases
- speech recognition
- end to end
- autoregressive
- scalable video
- rate adaptation
- non stationary
- hidden markov models
- language model
- pattern recognition
- random fields
- speech signal
- congestion control
- speech recognition systems
- video streaming
- spectral analysis
- sar images
- transport protocol
- neural network
- parameter estimation
- motion estimation