CHiVE: Varying Prosody in Speech Synthesis with a Linguistically Driven Dynamic Hierarchical Conditional Variational Network.
Tom KenterVincent WanChun-an ChanRob ClarkJakub VitPublished in: ICML (2019)
Keyphrases
- speech synthesis
- speech recognition
- text to speech
- prosodic features
- vocal tract
- network model
- wireless sensor networks
- network structure
- computer networks
- data driven
- computer vision
- data mining
- language model
- optical flow
- image segmentation
- peer to peer
- complex networks
- network traffic
- hidden markov models
- network architecture
- network topologies
- natural language