VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Envuelva su texto en etiquetas SSML para un control preciso:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Etiquetas el modelo seleccionado entiende — haga clic para soltar uno en su texto donde sucede:
Este modelo lee texto plano, por lo que las etiquetas en línea son ignoradas. Para la emoción basada en etiquetas, cambie a un modelo expresivo como Orfeo o Bark.
Definir pronunciaciones personalizadas (palabra = pronunciación):
Acerca de VITS
VITS — Variational Inference with adversarial learning for end-to-end Text-to-Speech — was introduced by Jaehyeon Kim and collaborators in 2021 and became a foundational architecture for modern neural speech. Rather than the older two-stage pipeline, it synthesizes audio in a single parallel end-to-end pass, pairing a variational autoencoder with normalizing flows and a GAN-style adversarial training process to lift naturalness. At about 25M parameters and trained on ~585 hours, it produces natural prosody at fast inference speeds and supports multiple speakers. It serves as a solid general-purpose, free baseline and underpins many later models such as Piper and MeloTTS.
Lo mejor para: General-purpose text-to-speech with natural prosody
Examinar todo VITS vocesDe un vistazo
- Desarrollador
- Jaehyeon Kim et al.
- Licencia
- MIT
- Nivel
- free
- Velocidad
- fast
- Clonación de voz
- No
- Idiomas
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Máx. caracteres
- 2000