Video Dubbing Translator
Video Dubbing Translator is a local-first, open-source pipeline for automatically dubbing videos into other languages. It chains together several AI stages: WhisperX for speech recognition and word-level timestamping, a Google-based translation step for converting the transcript, VoxCPM2 for voice cloning and speech synthesis, and an optional LatentSync stage for lip-sync alignment so the speaker's mouth matches the new audio. The project is designed to run on the user's own machine rather than as a hosted service, so the source media, transcripts, and cloned voice data stay on local hardware. Users supply a video file, choose a target language, and the tool performs transcription, translation, voice synthesis, and optional lip-sync in sequence. Because it relies on open models and libraries, it is fully free to use and does not require a subscription, though heavy GPU resources are typically needed for reasonable processing times. The repository is published on GitHub and is oriented toward users comfortable installing Python/CUDA dependencies and configuring model weights.
- Category
- Video
- Pricing
- Free
Problem solved
Dubbing a video into another language normally requires either expensive commercial services or a fragmented manual workflow of transcription, translation, recording, and re-editing. This tool automates the whole chain locally, letting users produce translated voice tracks that preserve the original speaker's voice and optionally re-align lip movement, without uploading media to a third-party cloud.
Use cases
- Dubbing a YouTube video into another language while preserving the original speaker's voice
- Localizing tutorial or course recordings for international audiences
- Translating and dubbing interviews or documentary footage without uploading confidential media
- Adding a translated voice track to short-form social clips for reach in other markets
- Prototyping voice-cloning and lip-sync research pipelines on local GPUs
- Re-dubbing older content into new languages as an inexpensive alternative to studio dubbing
Audience
- Content Creators
- Developers
- Researchers
- Hobbyists
Pros
- Runs locally, keeping video, transcripts, and cloned voice data on the user's machine
- Free and open source with no subscription or per-minute dubbing fees
- Combines transcription, translation, voice cloning, and optional lip-sync in one pipeline
- Uses established components such as WhisperX and LatentSync
- Optional lip-sync stage improves realism over audio-only dubbing
Cons
- Requires a capable GPU and manual installation of Python/CUDA and model dependencies
- Translation relies on a Google translation step, which may need network access and can be less controllable than a local model
- Output quality depends heavily on the quality of the source audio and the cloned voice model
- No managed hosting, so users handle setup, troubleshooting, and updates themselves
- Limited documentation, community, and support relative to commercial dubbing platforms
Official website: https://github.com/kadirb4rut/video-dubbing-translator
Canonical LambdaBase page: https://www.lambdabase.com/tools/video-dubbing-translator