Improving Hybrid CTC/Attention End-to-end Speech Recognition with Pretrained Acoustic and Language Model.
Keqi DengSongjun CaoYike ZhangLong MaPublished in: CoRR (2021)
Keyphrases
- end to end
- speech recognition
- language model
- speech recognition systems
- speaker independent
- speech recognizers
- language modeling
- acoustic models
- speech synthesis
- automatic speech recognition
- probabilistic model
- information retrieval
- speech recognizer
- n gram
- query expansion
- document retrieval
- retrieval model
- test collection
- speech signal
- noisy environments
- mixture model
- word error rate
- handwriting recognition
- translation model
- pattern recognition
- relevance model
- speaker identification
- query terms
- training data
- computer vision
- neural network