Combining recurrent neural networks and factored language models during decoding of code-Switching speech.
Heike AdelDominic TelaarNgoc Thang VuKatrin KirchhoffTanja SchultzPublished in: INTERSPEECH (2014)
Keyphrases
- language model
- recurrent neural networks
- speech recognition
- word error rate
- language modeling
- probabilistic model
- retrieval model
- n gram
- spoken term detection
- feed forward
- neural network
- automatic speech recognition
- speech signal
- language modelling
- document retrieval
- information retrieval
- echo state networks
- reservoir computing
- ad hoc information retrieval
- recurrent networks
- out of vocabulary
- statistical language models
- context sensitive
- test collection
- vector space model
- query expansion
- retrieval effectiveness
- smoothing methods
- query terms
- cascade correlation
- language models for information retrieval
- artificial neural networks
- document ranking
- error rate
- hidden markov models
- machine learning