dub-studio
Free offline AI video dubbing studio for Windows — voice cloning, translation, subtitles & on-screen-text localization. 100% local, one native .exe, zero Python
What is dub-studio?
Free offline AI video dubbing studio for Windows — voice cloning, translation, subtitles & on-screen-text localization. 100% local, one native .exe, zero Python.
Overview of dub-studio
What it is
Dub Studio is a free, offline AI video dubbing studio for Windows. It re-voices any video into another language using voice cloning, translates captions, and localizes on-screen text directly on the frame. The entire pipeline runs 100% locally on the user's machine with a single native .exe (Rust + C++/CUDA), requiring no Python or cloud subscription. Optionally, heavy stages such as translation, vision, TTS, and transcription can be offloaded to the cloud via OpenRouter, with each engine independently selectable. Models, engines, CUDA/VC++ runtime, and ffmpeg are downloaded and installed by the app on first run. An NVIDIA GPU is the tested and recommended setup.
Key features
- Voice cloning — the original timbre is cloned and speaks the new language using the native Higgs Audio v3 engine (GGUF). Auto-cast by speaker or bring a custom voice.
- Speaker diarization — identifies who speaks and when (NVIDIA Nemotron 3 Diarization, up to 8 voices), assigning a distinct voice per speaker.
- Character casting (beta) — binds faces to speakers by co-occurrence, saves a casting profile for the whole series, and auto-applies to the next episode.
- Choice of ASR engine — transcribe with Parakeet-TDT (GPU, default) or Whisper (Purfview faster-whisper standalone, CPU), with selectable model size and quantization.
- On-screen text localization — OCR detects baked-in text (PP-OCR ONNX), blurs the original, and prints a localized title on top in a matched style.
- Live editor — edit transcript, voices, caption style, blur boxes, and titles with ~0.17 s/frame preview, every change visible instantly.
- 26 caption presets — karaoke, word-by-word, hormozi, neon, and more, rendered on the user's own frame (WYSIWYG).
Use cases
- Dubbing videos into another language while preserving the original speaker's voice.
- Creating voice-over translations with the original audio ducked underneath.
- Burning original-language or translated subtitles onto video.
- Localizing on-screen text (e.g., titles, brands) in the video frame.
- Generating clean diarized transcripts with per-speaker layout and karaoke play-along.
- Funny remix: rewriting the entire script based on a theme and re-dubbing.
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
The license is not stated in the repository metadata. The software is free and open source.
Based on the project's public README and repository metadata.