Wav2Lip (offline lip-sync)
Lip-syncOfflineNon-commercial
Offline generative lip-sync (visual dubbing) via Wav2Lip ONNX. Torch-free, CPU-capable; regenerates the mouth to match dubbed audio. Weights are research/non-commercial.
Install it from Subtitld
Settings Add-ons Wav2Lip (offline lip-sync) Install
Subtitld downloads it, checks its SHA-256, and keeps it updated.
- Add-on download
- 176–274 MB
Options
In Settings › Add-ons › Wav2Lip (offline lip-sync) › Configure.
Mouth box padding (top bottom left right)
Default:
0 10 0 0Extra pixels added around the detected face box before lip-sync. Increase 'bottom' if the chin is cut off.
Resize factor
Default:
1Downscale each frame by this factor before processing. Higher = faster and lower resolution.
Face detection threshold
Default:
0.3SCRFD face-detector confidence threshold. Lower detects more (and weaker) faces.
Platforms
| System | Download |
|---|---|
| Linux x86-64 | 273.6 MB |
| macOS (Apple silicon) | 202.4 MB |
| Windows x86-64 | 175.9 MB |
License
Add-on
MIT (wrapper); Wav2Lip weights are RESEARCH/NON-COMMERCIAL only
MIT (wrapper); Wav2Lip weights research/non-commercial only
Versions
- 0.1.0