Prediction of Prosodic Word Boundaries in Chinese TTS Based on Maximum Entropy Markov Model and Transformation Based Learning.
Ziping ZhaoXirong MaPublished in: CIS (2012)
Keyphrases
- transformation based learning
- maximum entropy
- markov model
- prosodic features
- markov models
- text to speech
- speech synthesis
- maximum entropy principle
- statistical model
- markov chain
- speaker verification
- hidden markov models
- named entity recognition
- markov networks
- minimum cross entropy
- conditional random fields
- n gram
- co occurrence
- machine learning
- dynamic programming