Comparison of language models trained on written texts and speech transcripts in the context of automatic speech recognition.
Sebastian DziadzioAleksandra NaboznyAleksander Smywinski-PohlBartosz ZiólkoPublished in: FedCSIS (2015)
Keyphrases
- language model
- automatic speech recognition
- speech recognition
- word error rate
- broadcast news
- language modeling
- spoken term detection
- speech transcripts
- probabilistic model
- n gram
- information retrieval
- speech signal
- document retrieval
- error rate
- out of vocabulary
- retrieval model
- query expansion
- language modelling
- question answering
- hidden markov models
- test collection
- query terms
- translation model
- neural network
- acoustic features
- word recognition
- relevance model
- pseudo relevance feedback
- natural language
- pattern recognition