This repository provides a quickstart pipeline for generating a talking avatar video by syncing an input audio file with a static face image using deep learning–based lip synchronization.
The quickstart.py script is designed to help you run the full pipeline with minimal setup.
- Takes an input face image
- Takes an input audio file
- Uses a lip-sync model (Wav2Lip-based workflow)
- Generates a synchronized talking avatar video
avatar_sync/
- quickstart.py # Entry point (this file)
- wav2lip_workspace/ # Model & inference workspace
- requirements.txt # Python dependencies
- .gitignore