renjunok/infinitetalk2

1---2title: InfiniteTalk - Talking Video Generator3emoji: 🎬4colorFrom: blue5colorTo: purple6sdk: gradio7sdk_version: 5.6.08python_version: "3.10"9app_file: app.py10pinned: false11license: apache-2.012short_description: AI talking video generator with accurate lip-sync13---14 15# InfiniteTalk - Talking Video Generator16 17Generate realistic talking head videos with accurate lip-sync from images or dub existing videos with new audio!18 19## Features20 21- **Image-to-Video**: Transform a static portrait image into a talking video using audio input22- **Video Dubbing**: Re-sync an existing video with new audio while maintaining natural head movements and expressions23- **High Quality**: 480p and 720p resolution support with advanced lip-sync technology24- **Unlimited Length**: Support for videos of any duration through chunked processing25 26## How It Works27 28InfiniteTalk uses the state-of-the-art Wan2.1 diffusion model combined with specialized audio conditioning to create photorealistic talking videos. The system synchronizes:29 30- Lip movements with audio31- Head pose and rotations32- Facial expressions33- Body posture34 35## Usage36 37### Image-to-Video381. Upload a portrait image (clear face visibility recommended)392. Upload an audio file or use the example403. Adjust parameters if needed414. Click Generate42 43### Video Dubbing441. Upload a video with a visible face452. Upload new audio to dub over it463. Adjust parameters if needed474. Click Generate48 49## Parameters50 51- **Resolution**: Choose between 480p (faster) or 720p (higher quality)52- **Diffusion Steps**: More steps = higher quality but slower (20-50 recommended)53- **Audio Guide Scale**: Controls audio influence on generation (2-4 recommended)54- **Seed**: For reproducible results55 56## Credits57 58Built on [InfiniteTalk](https://github.com/MeiGen-AI/InfiniteTalk) by MeiGen-AI.59 60## License61 62Apache 2.0 - See LICENSE.txt for details63