Phoneme segmentation using deep learning for speech synthesis.
Young Han LeeJong-Yeol YangChoong Sang ChoHyedong JungPublished in: RACS (2018)
Keyphrases
- speech synthesis
- deep learning
- speech recognition
- text to speech
- vocal tract
- unsupervised learning
- prosodic features
- image segmentation
- machine learning
- unsupervised feature learning
- segmentation algorithm
- weakly supervised
- level set
- segmentation method
- multiscale
- mean shift
- mental models
- object class
- object segmentation
- shape prior
- object categories
- markov random field
- co occurrence
- hidden markov models
- active learning
- learning algorithm