VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Oorvloei jou teks in SSML etiket vir presiese beheer:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Merk die gekose model verstaan ooit die woord ooit om een in jou teks te laat val waar dit gebeur:
Hierdie model lees gewone teks, so inlyn etiket word geignoreer. Vir etiket-gebaseerde emosie, wissel na 'n uitdrukkingende model soos Orpheus of Bark.
Definieer pasmaak uitspraak (woord = uitspraak):
Aangaande 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.
Beste vir: General-purpose text-to-speech with natural prosody
Blaai deur almal VITS stemmeMet'n blik
- Ontwikkelingvloeistof is minDeveloper
- Jaehyeon Kim et al.
- Lisensie
- MIT
- Tier
- free
- Spoed
- fast
- Stem kloning
- Nee
- Tale
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Voeg- agteraan- by Taal
- 2000