JDI-T: Jointly trained Duration Informed Transformer for Text-To-Speech without Explicit Alignment.
Dan LimWon JangGyeonghwan OHyeyeong ParkBongwan KimJaesam YoonPublished in: CoRR (2020)
Keyphrases
- text to speech
- speech synthesis
- fuzzy logic
- prosodic features
- programming tool
- text to speech synthesis
- word processing
- training set
- dynamic time warping
- fault diagnosis
- power transformers
- power system
- multilayer perceptron
- english text
- image alignment
- distribution network
- svm classifier
- multi modal
- learning algorithm