Video Dubbing Translator
Local-first AI video dubbing and voice cloning
Details
- Follow on
- Categories
- AIDeveloper ToolsAudio & Music
- Target Audience
- DevelopersContent CreatorsAI Engineers
- Pricing
- Free
- Alternatives
ElevenLabs alternatives
Discovery signals
How AI and people discover Video Dubbing Translator on PeerPush
About Video Dubbing Translator
Video Dubbing Translator is an open-source, local-first Python app that translates videos while preserving the original speaker’s voice. It separates vocals and background audio, transcribes and aligns speech with Whisper/WhisperX, translates speech segments, generates 48 kHz cloned speech with VoxCPM2, fits the generated speech to the original timing, and reconstructs the video with the original background track. Optional LatentSync adds AI lip-sync on CUDA GPUs. The current UI exposes 11 validated dubbing target languages, while VoxCPM2 itself supports 30 languages. Media processing and AI inference run locally; the current translation step uses Google Translate via deep-translator. The project includes a browser GUI, CLI, setup/preflight tools, and a real English → Turkish comparison demo.
Screenshots
Reviews (1)
Average 5.0 out of 5
Based on 1 review
A promising local-first dubbing tool with a clear privacy angle and an interesting focus on practical voice workflows.
Comments (2)
A local-first dubbing workflow is a thoughtful direction; keeping voice processing close to the user could help with privacy and control.
@nckbrz Thanks! Privacy and user control are exactly why I chose a local-first approach. I’m also working on improving voice similarity, timing, and multilingual quality. Appreciate the feedback!
Built this as a local-first open-source dubbing pipeline. I’d love feedback on voice similarity, timing, multilingual quality, and CUDA performance. There’s an English→Turkish demo in the repo.