Using Dependency Grammar Features in Whole Sentence Maximum Entropy Language Model for Speech Recognition.
Teemu RuokolainenTanel AlumäeMarcus DobrinkatPublished in: Baltic HLT (2010)
Keyphrases
- speech recognition
- language model
- maximum entropy
- dependency structure
- speech recognition systems
- language modeling
- n gram
- hidden markov models
- document retrieval
- probabilistic model
- speech signal
- natural language
- information retrieval
- automatic speech recognition
- conditional random fields
- word error rate
- pattern recognition
- feature vectors
- query expansion
- prior knowledge
- machine learning
- test collection
- mixture model
- handwriting recognition
- feature extraction
- feature space
- named entity recognition
- feature set
- part of speech
- classification accuracy