Sentence Selection Based on Extended Entropy Using Phonetic and Prosodic Contexts for Statistical Parametric Speech Synthesis.
Takashi NoseYusuke AraoTakao KobayashiKomei SugiuraYoshinori ShigaPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2017)
Keyphrases
- speech synthesis
- speech recognition
- prosodic features
- information theoretic
- text to speech
- information theory
- hidden markov models
- vocal tract
- language model
- speech signal
- mutual information
- noisy environments
- natural language
- pattern recognition
- selection algorithm
- automatic speech recognition
- information entropy
- statistical inference
- speech corpus
- computer vision
- statistical methods
- image coding
- bayesian networks