harlanhong/ACTalker
An ICCV 2025 end-to-end video diffusion framework for talking head synthesis supporting audio and expression control.
Not currently ranked — collecting fresh signals.
star history
ACTalker is a video diffusion model for generating realistic talking head videos from audio and expression signals. It employs masked selective state spaces modeling to achieve audio-visual synchronization and natural motion in avatar animation. The framework supports both single and multi-signal control for digital human synthesis and is intended for high-quality video generation of talking avatars.
Frequently asked
- What is harlanhong/ACTalker?
- An ICCV 2025 end-to-end video diffusion framework for talking head synthesis supporting audio and expression control.
- Is ACTalker open source?
- Yes — harlanhong/ACTalker is an open-source project tracked on heatdrop.
- What language is ACTalker written in?
- harlanhong/ACTalker is primarily written in Python.
- How popular is ACTalker?
- harlanhong/ACTalker has 459 stars on GitHub.
- Where can I find ACTalker?
- harlanhong/ACTalker is on GitHub at https://github.com/harlanhong/ACTalker.