VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Kääri tekstisi SSML-tunnisteisiin tarkan valvonnan varmistamiseksi:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Merkitse valittu malli ymmärtää – klikkaa pudottaaksesi yhden tekstiisi, missä se tapahtuu:
Tässä mallissa lukee pelkkää tekstiä, joten rivitunnisteet jätetään huomiotta. Jos haluat tag-pohjaisen tunteen, vaihda ilmaisukykyiseen malliin, kuten Orfeus tai Bark.
Määrittele custom-äänet (sana = ääntäminen):
Tietoja 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.
Paras: General-purpose text-to-speech with natural prosody
Selaa kaikkia VITS äänetSilmäyksellä
- Kehittäjä
- Jaehyeon Kim et al.
- Lisenssi
- MIT
- Määrittämistaso
- free
- Nopeus
- fast
- Äänien kloonaus
- Ei tarvitse.
- Kielet
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Maksimimerkit
- 2000