VITS ТТС
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Матнни аниқ назорат учун SSML теглар билан ўраб қўйиш:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Танланган модел тушунадиган тегилар — уларни матнга тушириш учун босинг:
Бу модел оддий матн ўқийди, шунинг учун тегишли теги эътиборга олинмайди. Эмоционал теги асосида, Orpheus ёки Bark каби ифодали моделга ўтинг.
Ўз нутқини белгилаш (сўз = нутқ):
Маълумот 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.
Энг яхшиси: General-purpose text-to-speech with natural prosody
Ҳаммасини кўриш VITS овозларБир қарашда
- Ижодкор
- Jaehyeon Kim et al.
- Лицензия
- MIT
- Тир
- free
- Тезлик
- fast
- Овозни клонлаш
- Йўқ
- Тиллар
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Максимум ҳарфлар
- 2000