Prediction of Prosodic Phrase Boundaries in Chinese TTS Based on Conditional Random Fields and Transformation Based Learning.
Ziping ZhaoYaoting ZhuPublished in: FSKD (2) (2009)
Keyphrases
- conditional random fields
- transformation based learning
- maximum entropy
- text to speech
- graphical models
- prediction accuracy
- web page prediction
- higher order
- markov random field
- sequence labeling
- prosodic features
- random fields
- hidden markov models
- markov models
- named entity recognition
- generative model
- information extraction
- probabilistic model
- pairwise
- crf model
- image labeling
- markov networks
- semi markov
- speech synthesis
- structured learning
- parameter learning
- superpixels
- information retrieval
- segmentation method
- structured prediction
- graph cuts