DQR-TTS: Semi-supervised Text-to-speech Synthesis with Dynamic Quantized Representation.
Jiangzong WangPengcheng LiXulong ZhangNing ChengJing XiaoPublished in: CoRR (2023)
Keyphrases
- text to speech
- text to speech synthesis
- semi supervised
- image representation
- co training
- semi supervised learning
- dynamic environments
- labeled data
- supervised learning
- active learning
- training data
- image classification
- data sets
- probabilistic model
- hidden markov models
- feature representation
- labeled examples
- semi supervised classification
- prosodic features
- three dimensional