MegaTtsStage

Enums · AiDotNet.TextToSpeech.Latest

Which of Mega-TTS’s two models MegaTTS trains.

Fields

Field Summary
MegaTtsStage Acoustic The VQGAN-based acoustic model: the encoders, the duration predictor and the mel decoder against the multi-window discriminator (Jiang et al.
MegaTtsStage ProsodyLanguageModel The prosody large language model, on the frozen acoustic model’s prosody codes (§3.2).
int value__