Diff2Lip: Audio Conditioned Diffusion Models for Lip-Synchronization

TL;DR

Diff2Lip is a diffusion-based model that achieves high-quality lip synchronization in-the-wild, outperforming previous methods in image quality and synchronization accuracy by leveraging complete contextual information.

Contribution

This paper introduces Diff2Lip, a novel diffusion model for lip-sync that preserves identity and image quality, trained on in-the-wild datasets, and surpasses existing methods in key metrics.

Findings

Outperforms Wav2Lip and PC-AVS in FID and MOS scores

Effective in both reconstruction and cross settings

Operates successfully on in-the-wild videos

Abstract

The task of lip synchronization (lip-sync) seeks to match the lips of human faces with different audio. It has various applications in the film industry as well as for creating virtual avatars and for video conferencing. This is a challenging problem as one needs to simultaneously introduce detailed, realistic lip movements while preserving the identity, pose, emotions, and image quality. Many of the previous methods trying to solve this problem suffer from image quality degradation due to a lack of complete contextual information. In this paper, we present Diff2Lip, an audio-conditioned diffusion-based model which is able to do lip synchronization in-the-wild while preserving these qualities. We train our model on Voxceleb2, a video dataset containing in-the-wild talking face videos. Extensive studies show that our method outperforms popular methods like Wav2Lip and PC-AVS in Fr\'echet inception distance (FID) metric and Mean Opinion Scores (MOS) of the users. We show results on both reconstruction (same audio-video inputs) as well as cross (different audio-video inputs) settings on Voxceleb2 and LRW datasets. Video results and code can be accessed from our project page ( https://soumik-kanad.github.io/diff2lip ).