Non-autoregressive Transformer with Unified Bidirectional Decoder for Automatic Speech Recognition.
Chuan-Fei ZhangYan LiuTian-Hao ZhangSong-Lu ChenFeng ChenXu-Cheng YinPublished in: CoRR (2021)
Keyphrases
- autoregressive
- automatic speech recognition
- speech recognition
- non stationary
- moving average
- speech signal
- broadcast news
- gaussian markov random field
- hidden markov models
- word error rate
- random fields
- conversational speech
- spoken words
- speech retrieval
- sar images
- fault diagnosis
- recognition errors
- probabilistic model
- random field models
- speech corpus
- speech sounds
- linear prediction
- multiscale
- information retrieval
- machine learning