StyleTtsZsTrainingStage
Enums · AiDotNet.TextToSpeech.StyleEmotion
The modules a StyleTTS-ZS training step updates (Li et al. 2024, §III and §IV-A).
Fields
| Field | Summary |
|---|---|
StyleTtsZsTrainingStage Distillation |
Distillation of the denoiser’s guided DDIM sampling into the one-step student (Eq. |
StyleTtsZsTrainingStage Training |
The acoustic synthesizer, the prosody autoencoder, the style diffusion denoiser and the student’s pre-training (predicting the time-varying style at σ = 0 from zero noise), against the discriminators. |
int value__ |