Shared models
Models from https://huggingface.co/camenduru/Wav2Lip are shared with sharepoint, making it problematic to download into Colab. The models are shared here for convinience.
| Model | Description | Link to the model |
|---|---|---|
| Wav2Lip | Highly accurate lip-sync | wav2lip.pth |
| Wav2Lip + GAN | Slightly inferior lip-sync, but better visual quality | wav2lip_gan.pth |
| Expert Discriminator | Weights of the expert discriminator | lipsync_expert.pth |
| Visual Quality Discriminator | Weights of the visual disc trained in a GAN setup | visual_quality_disc.pth |