Phoneme HMMs constrained by frame correlations.
Satoshi TakahashiTatsuo MatsuokaYasuhiro MinamiKiyohiro ShikanoPublished in: ICASSP (2) (1993)
Keyphrases
- hidden markov models
- speech recognition
- phoneme recognition
- automatic speech recognition
- data sets
- video frames
- markov models
- speaker dependent
- pattern recognition
- frame rate
- conditional random fields
- temporal correlation
- chinese named entity recognition
- speech synthesis
- hidden states
- spatial correlation
- image frames
- image sequences
- machine learning