VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Ukufaka umbhalo wakho kumathegi we-SSML ukulawula okucacile:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Amathegi amamodeli akhethiwe aqonda - chofoza ukuwasusa kusihloko sakho lapho kwenzeka khona:
Le modeli ifunda umbhalo ojwayelekile, ngakho amathegi e-inline akhohlwa. Ukwenza umbono osekelwe kumathegi, shintsha kwimodeli ebonisa umbono njenge-Orpheus noma i-Bark.
Chaza ukuchaza okujwayelekile (igama = ukuchaza):
Ngo 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.
Okungcono kakhulu: General-purpose text-to-speech with natural prosody
Khangela konke VITS izizwiNgombono ocacile
- Umthuthukisi
- Jaehyeon Kim et al.
- Ilayisense
- MIT
- I-Tiger
- free
- Isivinini
- fast
- Ukuklona umsindo
- Akukho
- Izilimi
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Amaphawu aphezulu
- 2000