NAST: Noise Aware Speech Tokenization for Speech Language Models.
Shoval MessicaYossi AdiPublished in: CoRR (2024)
Keyphrases
- language model
- speech recognition
- word error rate
- automatic speech recognition
- noisy environments
- n gram
- speech signal
- language modeling
- spoken term detection
- probabilistic model
- document retrieval
- test collection
- audio visual
- information retrieval
- hidden markov models
- retrieval model
- context sensitive
- error rate
- statistical language models
- pseudo relevance feedback
- query expansion
- vector space model
- multi modal
- handwriting recognition
- document ranking
- broadcast news
- out of vocabulary
- language modelling
- document length
- relevance model
- ad hoc information retrieval
- language models for information retrieval
- query terms