# OmniSync: Towards Universal Lip Synchronization via Diffusion Transformers

Source: [OmniSync: Towards Universal Lip Synchronization via Diffusion Transformers](<https://www.sota2.com/research/paper/68fa7b5f1e209131cea87fd2>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T07:02:34.900Z

Authors: Ziqiao Peng, Jiwen Liu, Haoxian Zhang, Xiaoqiang Liu, Songlin Tang, Pengfei Wan, Di Zhang, Hongyan Liu, Jun He

Published: 2025-05-27T17:20:38.000Z

## Abstract

Lip synchronization is the task of aligning a speaker's lip movements in video with corresponding speech audio, and it is essential for creating realistic, expressive video content. However, existing methods often rely on reference frames and masked-frame inpainting, which limit their robustness to identity consistency, pose variations, facial occlusions, and stylized content. In addition, since audio signals provide weaker conditioning than visual cues, lip shape leakage from the original video will affect lip sync quality. In this paper, we present OmniSync, a universal lip synchronization framework for diverse visual scenarios. Our approach introduces a mask-free training paradigm using Diffusion Transformer models for direct frame editing without explicit masks, enabling unlimited-duration inference while maintaining natural facial dynamics and preserving character identity. During inference, we propose a flow-matching-based progressive noise initialization to ensure pose and identity consistency, while allowing precise mouth-region editing. To address the weak conditioning signal of audio, we develop a Dynamic Spatiotemporal Classifier-Free Guidance (DS-CFG) mechanism that adaptively adjusts guidance strength over time and space. We also establish the AIGC-LipSync Benchmark, the first evaluation suite for lip synchronization in diverse AI-generated videos. Extensive experiments demonstrate that OmniSync significantly outperforms prior methods in both visual quality and lip sync accuracy, achieving superior results in both real-world and AI-generated videos.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Lip Synchronization on HDTF](<https://www.sota2.com/research/sota/lip-synchronization-on-hdtf.md>): Lip synchronization; HDTF; FID: 7.855.

- [Talking Head Generation on RealWorld-LipSync](<https://www.sota2.com/research/sota/talking-head-generation-on-realworld-lipsync.md>): Talking Head Generation; RealWorld-LipSync; FID: 13.655.

- [Lip Synchronization on AIGC-LipSync](<https://www.sota2.com/research/sota/lip-synchronization-on-aigc-lipsync.md>): Lip synchronization; AIGC-LipSync; FID: 10.681.

- [Video Dubbing on RealWorld-LipSync](<https://www.sota2.com/research/sota/video-dubbing-on-realworld-lipsync.md>): Video Dubbing; RealWorld-LipSync; Lip Sync Accuracy: 3.37.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2505.21448>)

[PDF](<https://arxiv.org/pdf/2505.21448.pdf>)