A segmental CRF approach to large vocabulary continuous speech recognition.
Geoffrey ZweigPatrick NguyenPublished in: ASRU (2009)
Keyphrases
- conditional random fields
- hidden markov models
- automatic speech recognition
- crf model
- pairwise
- maximum entropy
- sequence labeling
- markov random field
- information extraction
- graphical models
- semantic segmentation
- random fields
- speech recognizer
- speech recognition
- superpixels
- sequence alignment
- segmentation method
- state transition
- parameter learning
- non stationary
- machine learning
- real world
- neural network