VITS TTS
The end-to-end TTS architecture that combines a variational autoencoder, normalizing flows, and adversarial training.
SSML tags ሒዝካ ጽሑፍካ ሒዝካ ንምውሳድ:
<speak><prosody rate="slow">Slow speech</prosody></speak>
ርኢቶታት እቲ ዝተመርጸ ሞዴል ዝፈልጦ — ጠቅልል ንኸውዕሎ ኣብ ጽሑፍካ ኣብ ዝግበር ቦታ:
እዚ ሞዴል'ዚ ጽሑፍ ቀሊል ይንብብ፣ ከምኡ'ውን ኣብ መስመር ዝርከብ ቴግታት ይቕረ ይበሃል። ን tag-based emotion፣ ናብ ሞዴል ስነ-ኣእምሮኣዊ ከም Orpheus ወይ Bark ምቕያር
ድምጺ ተለፎን (ቓል = ድምጺ)
ብዛዕባ 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.
ምርኣይ: General-purpose text-to-speech with natural prosody
ርአ VITS ቃላትኣብ ሓደ ገጽ
- መተግበሪያ
- Jaehyeon Kim et al.
- ውልቀ-መዚ
- MIT
- ቍጽሪ
- free
- ፍጥነት
- fast
- ድምጺ
- ኣይ
- ቋንቋ
- English, German, Spanish, French, Portuguese, Dutch, Finnish, Hungarian, Bulgarian, Japanese, Polish
- ቍጽሪ ኣርእስቲ
- 2000