Document-based Dirichlet class language model for speech recognition using document-based n-gram events.
Md. Akmal HaidarDouglas D. O'ShaughnessyPublished in: SLT (2014)
Keyphrases
- language model
- n gram
- document retrieval
- document ranking
- document length
- speech recognition
- ad hoc information retrieval
- language modeling
- document representation
- vector space model
- information retrieval
- query terms
- language modelling
- probabilistic model
- web documents
- query specific
- relevance model
- mixture model
- retrieval model
- word clouds
- word level
- dirichlet prior
- test collection
- language independent
- text classification
- document images
- context sensitive
- language modeling framework
- query expansion
- out of vocabulary
- bag of words
- document analysis
- document collections
- document clustering
- part of speech
- relevant documents
- information retrieval systems
- text retrieval
- pseudo relevance feedback
- retrieval systems
- tf idf
- statistical language modeling
- translation model
- text classifiers
- retrieved documents
- term dependencies
- cross lingual