SpeechForensics: Audio-Visual Speech Representation Learning for Face Forgery Detection

TL;DR

This paper introduces a novel audio-visual speech representation learning approach for face forgery detection, significantly improving cross-dataset generalization and robustness without using fake videos during training.

Contribution

It proposes a self-supervised learning method that encodes audio-visual speech features for detecting face forgeries, enhancing generalization to unseen datasets.

Findings

Outperforms state-of-the-art methods in cross-dataset tests.

Achieves robustness against common perturbations.

Does not require fake videos for training.

Abstract

Detection of face forgery videos remains a formidable challenge in the field of digital forensics, especially the generalization to unseen datasets and common perturbations. In this paper, we tackle this issue by leveraging the synergy between audio and visual speech elements, embarking on a novel approach through audio-visual speech representation learning. Our work is motivated by the finding that audio signals, enriched with speech content, can provide precise information effectively reflecting facial movements. To this end, we first learn precise audio-visual speech representations on real videos via a self-supervised masked prediction task, which encodes both local and global semantic information simultaneously. Then, the derived model is directly transferred to the forgery detection task. Extensive experiments demonstrate that our method outperforms the state-of-the-art methods in terms of cross-dataset generalization and robustness, without the participation of any fake video in model training. Code is available at https://github.com/Eleven4AI/SpeechForensics.