A Multimodal Approach to Device-Directed Speech Detection with Large Language Models.
Dominik WagnerAlexander W. ChurchillSiddharth SigtiaPanayiotis G. GeorgiouMatt MirsamadiAarshee MishraErik MarchiPublished in: ICASSP (2024)
Keyphrases
- language model
- speech recognition
- language modeling
- word error rate
- document retrieval
- information retrieval
- n gram
- probabilistic model
- retrieval model
- audio visual
- query expansion
- mixture model
- spoken term detection
- language modelling
- statistical language models
- context sensitive
- speech signal
- automatic speech recognition
- test collection
- language model for information retrieval
- pseudo relevance feedback
- vector space model
- query terms
- multi modal
- document ranking
- smoothing methods
- feature selection
- term dependencies
- cross lingual
- error rate