Modeling heterogeneous data sources for speech recognition using synchronous hidden Markov models.
Yong ZhaoBiing-Hwang JuangPublished in: ICASSP (2013)
Keyphrases
- hidden markov models
- speech recognition
- heterogeneous data sources
- speech signal
- automatic speech recognition
- phoneme recognition
- data sources
- markov models
- sequential data
- gesture recognition
- speaker independent
- automatic speech recognition systems
- sequence classification
- text recognition
- conditional random fields
- data integration
- markov model
- audio signals
- keyword spotting
- speech recognizer
- visual speech recognition
- database systems
- discriminative training
- viterbi algorithm
- databases
- acoustic features
- broadcast news
- noisy environments
- acoustic models
- heterogeneous data
- pattern recognition