Hybrid HMM/BN LVCSR system integrating multiple acoustic features.
Konstantin MarkovSatoshi NakamuraPublished in: ICASSP (1) (2003)
Keyphrases
- integrating multiple
- acoustic features
- speech signal
- automatic speech recognition
- hidden markov models
- speech recognition
- bayesian networks
- music genre classification
- music information retrieval
- visual features
- speaker verification
- finite state transducers
- visual speech
- environmental sounds
- neural network
- audio features
- multi modal
- collaborative filtering
- feature extraction
- information retrieval