Fully-Hierarchical Fine-Grained Prosody Modeling For Interpretable Speech Synthesis.
Guangzhi SunYu ZhangRon J. WeissYuan CaoHeiga ZenYonghui WuPublished in: ICASSP (2020)
Keyphrases
- fine grained
- speech synthesis
- speech recognition
- text to speech
- coarse grained
- prosodic features
- tightly coupled
- vocal tract
- speech corpus
- access control
- high level
- language model
- graphical models
- co occurrence
- massively parallel
- automatic speech recognition
- probabilistic model
- bayesian networks
- web services
- knowledge base
- machine learning
- database