Rudrabha/Wav2Lip
Source: https://github.com/rudrabha/wav2lip
Structured fields
{
"archived": false,
"checked_at": "2026-10-17T00:00:00.000Z",
"usefulness": "Original research code (ACM Multimedia 2020) for lip-syncing an existing video to arbitrary audio; the ancestor of many later lip-sync tools. Restrictive use terms make it a research/non-commercial reference.",
"source_type": "repository",
"availability": "available",
"dependencies": [
"Python 3.6",
"ffmpeg",
"s3fd face-detection weights"
],
"next_check_at": "2026-10-31T00:00:00.000Z",
"known_limitations": [
"Python 3.6-era dependencies; low native face resolution relative to newer methods (inference).",
"Maintenance status/last commit date not visible on the fetched page."
],
"upstream_updated_at": "2025-06-22T02:41:21.000Z"
}Evidence claims
- [upstream_claim] README: lip-syncs videos to a target audio for any identity, voice and language; models trained on LRS2. source
- [sourced_fact] README states the code and models can only be used for personal/research/non-commercial purposes and that any commercial use is strictly prohibited; it points commercial users to a separate hosted service. source
- [upstream_claim] Pretrained Wav2Lip and Wav2Lip+GAN checkpoints are linked from Google Drive; face-detection weights (s3fd.pth) must be downloaded separately; Python 3.6 and ffmpeg required. source
- [upstream_claim] Research review verdict at export: reviewed (as of 2026-10-17). Recorded by the research pipeline, not a hub review outcome; evidence.review stays unreviewed until an independent reviewer records one.
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| code | unknown | restricted | restricted | restricted | unknown | README restricts use to personal/research/non-commercial purposes and names no standard open-source license; a LICENSE file was not found at the master path tried. |
| model_weights | unknown | restricted | restricted | restricted | unknown | Disclaimer says models are trained on LRS2 and commercial use is strictly prohibited. |