R3gm/SoniTranslate

Synchronized Translation for Videos. Video dubbing

57
/ 100
Established

Extracts audio from videos using speaker diarization (Pyannote), translates transcribed content across 100+ languages, and synthesizes new audio with voice cloning to match original speaker characteristics. Built on Gradio for the web interface, it integrates Hugging Face models for speech recognition and generation, with GPU acceleration via CUDA for real-time processing. Deployable locally, on Google Colab, or via Hugging Face Spaces for browser-based video dubbing without manual subtitle editing.

1,341 stars.

No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

1,341

Forks

320

Language

Python

License

Apache-2.0

Last pushed

Dec 02, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/R3gm/SoniTranslate"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.