Discriminative duration modeling for speech recognition with segmental conditional random fields.
Justine T. KaoGeoffrey ZweigPatrick NguyenPublished in: ICASSP (2011)
Keyphrases
- speech recognition
- hidden markov models
- conditional random fields
- crf model
- discriminative training
- sequence labeling
- random fields
- speech recognizer
- automatic speech recognition
- maximum entropy
- structured prediction
- speech synthesis
- probabilistic model
- named entity recognition
- speech signal
- speech recognition systems
- discriminative models
- speaker independent
- speech recognition technology
- language model
- speaker identification
- graphical models
- unsupervised learning
- pattern recognition
- semi markov
- markov random field
- web page prediction
- noisy environments
- information extraction
- pairwise
- segmentation method
- generative model
- feature extraction
- image processing