VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Verpacken Sie Ihren Text in SSML-Tags für eine präzise Kontrolle:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Tags, die das ausgewählte Modell versteht — klicken Sie, um einen in Ihren Text zu legen, wo es passiert:
Dieses Modell liest Text, so dass Inline-Tags ignoriert werden. Für tag-basierte Emotion, wechseln Sie zu einem ausdrucksstarken Modell wie Orpheus oder Bark.
Benutzerdefinierte Aussprachen definieren (Wort = Aussprache):
Über 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.
Das Beste für: General-purpose text-to-speech with natural prosody
Alle durchsuchen VITS StimmenAuf einen Blick
- Entwickler
- Jaehyeon Kim et al.
- Lizenz
- MIT
- Tierart
- free
- Geschwindigkeit
- fast
- Klonen der Stimme
- Nein
- Sprachen
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Maximale Zeichen
- 2000