Gated Recurrent Fusion with Joint Training Framework for Robust End-to-End Speech Recognition.
Cunhang FanJiangyan YiJianhua TaoZhengkun TianBin LiuZhengqi WenPublished in: CoRR (2020)
Keyphrases
- end to end
- speech recognition
- language model
- noisy environments
- hidden markov models
- speech synthesis
- pattern recognition
- text localization and recognition
- speech signal
- isolated word
- speech recognizer
- congestion control
- speaker identification
- acoustic models
- training process
- speech processing
- neural network
- speech recognizers
- automatic speech recognition
- audio visual speech recognition
- wall street journal corpus