VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
صحيح ڪنٽرول لاءِ پنھنجو متن SSML ٽيگ ۾ ويڙھيو:
<speak><prosody rate="slow">Slow speech</prosody></speak>
ٽيگ جيڪي چونڊيل ماڊل سمجھي ٿو - هڪ کي پنھنجي متن ۾ جتي ٿئي ٿو ڦيريڻ لاءِ ڪلڪ ڪريو:
هي ماڊل عام متن پڙهندو آھي، تنھنڪري لاٽ ۾ ٽيگ نظرانداز ڪيا ويندا آھن. ٽيگ تي ٻڌل احساسن لاءِ، ھڪ اظهاري ماڊل وانگر Orpheus يا Bark تي تبديل ڪريو.
پنھنجو آواز بيان ڪريو (شيء = آواز):
بابت 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.
بهترين: General-purpose text-to-speech with natural prosody
سڀ لکو VITS آوازهڪ نظر ۾
- ڊيولپر
- Jaehyeon Kim et al.
- لائسنس
- MIT
- جانور
- free
- رفتار
- fast
- آواز جو کلون
- نه
- ٻوليون
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- وڌيڪ نشان
- 2000