VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Umotaj svoj tekst u SSML oznake za preciznu kontrolu:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Oznake odabrani model razumije — kliknite da bacite jedan u vaš tekst gdje se to događa:
Ovaj model čita običan tekst, tako da inline oznake se zanemaruju. Za tag-based emocije, prebaci na ekspresivni model kao što su Orfeus ili Bark.
Definiši vlastite izgovore (riječ = izgovor):
O 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.
Najbolje za: General-purpose text-to-speech with natural prosody
Pregledaj sve VITS glasoviNa jedan pogled
- Programer
- Jaehyeon Kim et al.
- Dozvola
- MIT
- Nivo
- free
- Brzina
- fast
- Kloniranje glasa
- Ne.
- Jezici
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Maks. znakova
- 2000