Knowledge Transfer and Distillation from Autoregressive to Non-Autoregessive Speech Recognition.
Xun GongZhikai ZhouYanmin QianPublished in: INTERSPEECH (2022)
Keyphrases
- knowledge transfer
- autoregressive
- speech recognition
- non stationary
- knowledge sharing
- hidden markov models
- transfer learning
- language model
- random fields
- linear prediction
- pattern recognition
- speech recognition systems
- sar images
- learning tasks
- speech signal
- graph cuts
- maximum entropy
- markov random field
- supervised learning
- multiscale
- training data
- learning algorithm