VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
Ngresiki teks ing tag SSML kanggo kontrol presisi:
<speak><prosody rate="slow">Slow speech</prosody></speak>
Tag kang dipahami model kang dipilih - klik kanggo ngethok siji ing teks sampeyan ing ngendi iku kedadeyan:
Model iki maca teks biasa, mula tag ing baris diabaikan. Kanggo emosi berbasis tag, ganti menyang model ekspresif kaya Orpheus utawa Bark.
Nyathet tembung-tembung standar (kata = tembung):
Ngendi 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.
Paling apik kanggo: General-purpose text-to-speech with natural prosody
Jajal kabeh VITS swaraIng cetha
- Pangembang
- Jaehyeon Kim et al.
- Lisénsi
- MIT
- Tanggal
- free
- Kecepatan
- fast
- Kloning swara
- Ora
- Basa
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- Aksara paling akèh
- 2000