GenSync: A Generalized Talking Head Framework for Audio-driven Multi-Subject Lip-Sync using 3D Gaussian Splatting
Year: 2025
Type: preprint
Abstract: We introduce GenSync, a novel framework for multi-identity lip-synced video synthesis using 3D Gaussian Splatting. Unlike most existing 3D methods that require training a new model for each identity , GenSync learns a unified network that synthesizes lip-synced videos for multiple speakers. By incorporating a Disentanglement Module, our approach separates identity-specific features from audio representations, enabling efficient multi-identity video synthesis. This design reduces computational overhead and achieves 6.8x faster training compared to state-of-the-art models, while maintaining high lip-sync accuracy and visual quality.
Source: arXiv (Cornell University)
Authors: Agarwal, Anushka, Hassan, Muhammad Yusuf, Chafekar, Talha
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4415029081
Cites:
Cited by:
Related to: