Context-dependent Label Smoothing Regularization for Attention-based End-to-End Code-Switching Speech Recognition.
Zheying HuangPeng LiJi XuPengyuan ZhangYonghong YanPublished in: ISCSLP (2021)
Keyphrases
- end to end
- speech recognition
- context dependent
- hidden markov models
- speech processing
- language model
- speech signal
- speech recognizer
- speech synthesis
- automatic speech recognition
- speech recognition technology
- noisy environments
- pattern recognition
- low level
- congestion control
- speaker identification
- natural language
- speech recognition systems
- high level
- computer vision
- smoothing algorithm
- machine learning