Leveraging Acoustic and Linguistic Embeddings from Pretrained Speech and Language Models for Intent Classification.
Bidisha SharmaMaulik C. MadhaviHaizhou LiPublished in: ICASSP (2021)
Keyphrases
- language model
- speech recognition
- language modeling
- word error rate
- n gram
- probabilistic model
- document retrieval
- language modelling
- classification accuracy
- pattern recognition
- information retrieval
- retrieval model
- machine learning
- query expansion
- feature selection
- statistical language models
- spoken term detection
- vector space
- text classification
- ad hoc information retrieval
- test collection
- context sensitive
- vector space model
- feature space
- natural language
- decision trees
- pseudo relevance feedback
- broadcast news
- smoothing methods
- feature extraction
- acoustic signal