Lightspeech: Lightweight Non-Autoregressive Multi-Speaker Text-To-Speech.
Song LiBeibei OuyangLin LiQingyang HongPublished in: SLT (2021)
Keyphrases
- lightweight
- autoregressive
- text to speech
- prosodic features
- speech synthesis
- moving average
- non stationary
- speaker verification
- gaussian markov random field
- random fields
- programming tool
- autoregressive model
- word processing
- text to speech synthesis
- speech recognition
- wireless sensor networks
- spectrum analysis
- sar images
- communication infrastructure
- random field models
- audio visual
- expectation maximization
- multiresolution