SoundStormTrainingStage
Enums · AiDotNet.TextToSpeech.CodecBased
Which of SoundStorm’s two networks a training step updates.
Fields
| Field | Summary |
|---|---|
SoundStormTrainingStage SemanticToAcoustic |
The SoundStorm acoustic model (§3): semantic tokens to the SoundStream codes. |
SoundStormTrainingStage TextToSemantic |
The ByT5 text-to-semantic model’s decoder (§5): text bytes to semantic tokens. |
int value__ |