Fitting long-range information using interpolated distanced n-grams and cache models into a latent dirichlet language model for speech recognition.
Md. Akmal HaidarDouglas D. O'ShaughnessyPublished in: INTERSPEECH (2013)
Keyphrases
- language model
- speech recognition
- n gram
- probabilistic model
- long range
- language modelling
- language modeling
- translation model
- information retrieval
- bag of words
- retrieval model
- test collection
- query expansion
- statistical language modeling
- speech signal
- mixture model
- automatic speech recognition
- document retrieval
- information extraction
- variable length
- relevance model
- information retrieval systems
- language independent
- neural network
- hidden markov models
- pattern recognition
- word error rate
- graphical models
- query terms
- web documents