Improving Non-Autoregressive End-to-End Speech Recognition with Pre-Trained Acoustic and Language Models.
Keqi DengZehui YangShinji WatanabeYosuke HiguchiGaofeng ChengPengyuan ZhangPublished in: ICASSP (2022)
Keyphrases
- end to end
- speech recognition
- language model
- autoregressive
- pre trained
- speech recognition systems
- speaker independent
- language modeling
- n gram
- non stationary
- probabilistic model
- random fields
- training data
- speech signal
- training examples
- automatic speech recognition
- information retrieval
- word error rate
- machine learning
- markov random field
- motion estimation
- active learning
- decision trees
- learning algorithm