ARoBERT: An ASR Robust Pre-Trained Language Model for Spoken Language Understanding.
Chengyu WangSuyang DaiYipeng WangFei YangMinghui QiuKehan ChenWei ZhouJun HuangPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2022)
Keyphrases
- language understanding
- language model
- speech recognition
- pre trained
- language modeling
- automatic speech recognition
- natural language understanding
- word error rate
- n gram
- probabilistic model
- language processing
- query expansion
- retrieval model
- information retrieval
- natural language
- semantic interpretation
- test collection
- speech signal
- spoken dialogue systems
- dialogue system
- general knowledge
- training data
- natural language processing
- support vector
- markov random field
- feature space
- cognitive psychology
- feature selection