VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
सटीक नियन्त्रणका लागि SSML ट्यागमा तपाईँको पाठ बेर्नुहोस्:
<speak><prosody rate="slow">Slow speech</prosody></speak>
ट्याग चयन गरिएको मोडेल बुझ्छ - यो कहाँ हुन्छ आफ्नो पाठ मा एक गिर गर्न क्लिक:
यो नमूनाले सादा पाठ पढ्दछ, त्यसैले इनलाइन ट्याग उपेक्षा गरिन्छ । ट्याग-आधारित भावनाका लागि, ओर्फिस वा बारक जस्तै अभिव्यक्तिमूलक नमूनामा स्विच गर्नुहोस् ।
अनुकूल उच्चारण परिभाषित गर्नुहोस् (शब्द = उच्चारण):
यसका बारेमा 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.
यसका लागि उत्तम: General-purpose text-to-speech with natural prosody
सबै ब्राउज गर्नुहोस् VITS आवाजएक नजरमा
- विकासकर्ता
- Jaehyeon Kim et al.
- इजाजतपत्र
- MIT
- टर
- free
- गति
- fast
- आवाज क्लोनिङ
- होइन
- भाषा
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- अधिकतम क्यारेक्टर
- 2000