Say Goodbye to RNN-T Loss: A Novel CIF-based Transducer Architecture for Automatic Speech Recognition.
Tian-Hao ZhangDinghao ZhouGuiping ZhongBaoxiang LiPublished in: CoRR (2023)
Keyphrases
- automatic speech recognition
- speech recognition
- speech signal
- hidden markov models
- broadcast news
- conversational speech
- recurrent neural networks
- nearest neighbor
- word error rate
- recognition errors
- word recognition
- spoken words
- speech retrieval
- spontaneous speech
- noisy environments
- retrieval systems
- speech corpus
- error rate
- natural images