PhemeTrainingStage
Enums · AiDotNet.TextToSpeech.CodecBased
Which of Pheme’s two networks a training step updates.
Fields
| Field | Summary |
|---|---|
PhemeTrainingStage SemanticToAcoustic |
The SoundStorm-style semantic-to-acoustic model: semantic tokens to the acoustic codebooks. |
PhemeTrainingStage TextToSemantic |
The T5 text-to-semantic model: phonemes to SpeechTokenizer semantic tokens. |
int value__ |