Token-wise Training for Attention Based End-to-end Speech Recognition.
Peidong WangJia CuiChao WengDong YuPublished in: ICASSP (2019)
Keyphrases
- end to end
- speech recognition
- wall street journal corpus
- isolated word
- hidden markov models
- speech processing
- pattern recognition
- acoustic models
- speech signal
- speech synthesis
- automatic speech recognition
- language model
- speech recognizer
- noisy environments
- congestion control
- speaker identification
- speech recognition systems
- speech recognition technology
- speaker dependent
- neural network
- feature space
- feature selection
- information retrieval