Discriminative training of n-gram classifiers for speech and text routing.
Ciprian ChelbaAlex AceroPublished in: INTERSPEECH (2003)
Keyphrases
- n gram
- discriminative training
- automatic speech recognition
- language model
- co training
- text classification
- speech recognition
- hidden markov models
- word level
- character n grams
- finite state transducers
- english text
- probabilistic interpretation
- text to speech
- web documents
- language independent
- language specific
- maximum likelihood
- information retrieval
- variable length
- word segmentation
- training data
- viterbi algorithm
- text classifiers
- language modeling
- part of speech
- support vector
- speech signal
- text mining
- out of vocabulary
- naive bayes
- feature selection
- semi supervised
- support vector machine
- text retrieval
- broadcast news
- class labels
- training examples
- decision trees
- machine translation
- text categorization
- probabilistic model
- high dimensional
- keywords