Multiple-hypothesis CTC-based semi-supervised adaptation of end-to-end speech recognition.
Cong-Thanh DoRama DoddipatlaThomas HainPublished in: CoRR (2021)
Keyphrases
- end to end
- speech recognition
- multiple hypothesis
- semi supervised
- speaker adaptation
- particle filter
- hidden markov models
- language model
- automatic speech recognition
- semi supervised learning
- speech signal
- speech processing
- active learning
- pattern recognition
- speech recognition technology
- speech recognizer
- labeled data
- noisy environments
- unlabeled data
- speaker identification
- congestion control
- speech synthesis
- speaker independent
- isolated word
- speech recognition systems
- machine learning
- co training