VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Daangeessii kitaaba keessan keessaa tag SSML akka itti fayyadamtan:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Tag'oota mo'ellaa filatamee beekuu - cuqaasi akka tokkotti galchiin gara teekstaatti yoo ta'e:
Mo'ellaan kun kitaaba salphaa baraa, kan akka taggaa inniin linjii hin beekkamne. Akka taggaa-based emooshiniitti, mo'ellaa akka Orpheus ykn Bark.
Haalli fuula
Fuulaa 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.
Fakkeenyaaf: General-purpose text-to-speech with natural prosody
Fuulaa VITS DhaamsaAkkasumas
- Deebi'aa
- Jaehyeon Kim et al.
- Lizenz
- MIT
- Daandiin
- free
- Jijjiiramni
- fast
- Dhaabbilee
- Haata'u
- Afaan Oromoo
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Akkasumas
- 2000