Structured Sparse Attention for end-to-end Automatic Speech Recognition.
Jiabin XueTieran ZhengJiqing HanPublished in: ICASSP (2020)
Keyphrases
- end to end
- automatic speech recognition
- speech recognition
- speech signal
- hidden markov models
- word error rate
- conversational speech
- speech retrieval
- broadcast news
- wireless ad hoc networks
- congestion control
- admission control
- acoustic features
- speech corpus
- noisy environments
- recognition errors
- spontaneous speech
- spoken words
- speaker adaptation
- word recognition
- internet protocol
- real world
- sparse representation
- speech synthesis
- language model
- compound words
- response time
- application layer