Non-Autoregressive ASR Modeling Using Pre-Trained Language Models for Chinese Speech Recognition.
Fu-Hao YuKuan-Yu ChenKe-Han LuPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2022)
Keyphrases
- speech recognition
- language model
- autoregressive
- automatic speech recognition
- pre trained
- speech synthesis
- language modeling
- word error rate
- probabilistic model
- retrieval model
- n gram
- random fields
- non stationary
- speech signal
- query expansion
- noisy environments
- test collection
- information retrieval
- pattern recognition
- hidden markov models
- handwriting recognition
- query terms
- word segmentation
- training examples