VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Pakua maandishi yako katika tovuti ya SSML kwa ajili ya udhibiti sahihi:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Tag anaelewa mfano unaochaguliwa na unajibu ujumbe huu:
Mfano huu unasomeka maandishi rahisi, kwa hiyo alama za vidole hupuuzwa. Kwa hisia za ndani za watu, geukia kigezo kinachoonesha hisia kama Orfeus au Bark.
Matamshi ya desturi (neno = matamshi):
Habari 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.
Bora kwa: General-purpose text-to-speech with natural prosody
Ng'ombe wote VITS sautiKutupia jicho
- Mbuni
- Jaehyeon Kim et al.
- Lenzi
- MIT
- Tier
- free
- Mwendo
- fast
- Kufanyizwa kwa Sauti
- Hapana
- Lugha
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Wahusika wa Max
- 2000