Comprehensive Transformer TTS Versions Save
A Non-Autoregressive Transformer based Text-to-Speech, supporting a family of SOTA transformers with supervised and unsupervised duration modelings. This project grows with the research community, aiming to achieve the ultimate TTS
v0.2.1
2 years agoFix and update codebase & pre-trained models with demo samples
- Fix variance adaptor to make it work with all combinations of building block and variance type/level
- Update pre-trained models with demo samples of LJSpeech and VCTK under "transformer_fs2" building block and "cwt" pitch conditioning
- Share the result of ablation studies of comparing "transformer" vs. "transformer_fs2" paired among three types of pitch conditioning ("frame", "ph", and "cwt")
v0.2.0
2 years agoA lot of improvements with new features!
- Prepare two different types of data pipeline in preprocessor to maximize unsupervised/supervised duration modelings
- Adopt wavelet for pitch modeling & loss
- Add fine-trained duration loss
- Apply
var_start_stepsfor better model convergence, especially under unsupervised duration modeling - Remove dependency of energy modeling on pitch variance
- Add "transformer_fs2" building block, which is more close to the original FastSpeech2 paper
- Add two types of prosody modeling methods
- Loss camparison on validation set:
- LJSpeech - blue: v0.1.1 / green: v0.2.0
- VCTK - skyblue: v0.1.1 / orange: v0.2.0
