VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Zabalte svůj text do značek SSML pro přesné ovládání:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Značky vybraného modelu rozumí? klikněte na tlačítko pro kapku jednoho do textu, kde se to stane:
Tento model čte prostý text, takže inline značky jsou ignorovány. Pro tag-based emotion, přepněte na expresivní model, jako je Orpheus nebo Bark.
Definovat vlastní výslovnosti (slovo = výslovnost):
O aplikaci 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.
Nejlepší pro: General-purpose text-to-speech with natural prosody
Procházet vše VITS hlasyNa první pohled
- Vývojář
- Jaehyeon Kim et al.
- Licence
- MIT
- Úroveň
- free
- Rychlost
- fast
- Klonování hlasu
- Ne.
- Jazyky
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Max znaků
- 2000