A Complementary Joint Training Approach Using Unpaired Speech and Text for Low-Resource Automatic Speech Recognition.
Ye-Qian DuJie ZhangQiu-Shi ZhuLi-Rong DaiMing-Hui WuXin FangZhou-Wang YangPublished in: CoRR (2022)
Keyphrases
- automatic speech recognition
- conversational speech
- speech recognition
- spontaneous speech
- discriminative training
- speech signal
- spoken words
- acoustic models
- word error rate
- hidden markov models
- broadcast news
- speech retrieval
- recognition errors
- information retrieval
- noisy environments
- word recognition
- speech corpus
- phoneme recognition
- acoustic features
- speech sounds
- speech synthesis
- noisy images
- text retrieval
- speech recognizers
- speaker independent
- text documents
- error rate
- speaker adaptation
- compound words
- word level