Learn from Real: Reality Defender's Submission to ASVspoof5 Challenge

TL;DR

This paper introduces Reality Defender's system for audio deepfake detection, utilizing a novel self-supervised pretraining approach that enhances generalizability across multiple datasets with low training costs.

Contribution

The paper presents a new pretraining strategy using style-linguistics embeddings learned via contrastive learning to improve deepfake detection robustness.

Findings

Achieved 5.5% EER on ASVspoof5

Outperformed previous methods on multiple datasets

Low computational cost during training

Abstract

Audio deepfake detection is crucial to combat the malicious use of AI-synthesized speech. Among many efforts undertaken by the community, the ASVspoof challenge has become one of the benchmarks to evaluate the generalizability and robustness of detection models. In this paper, we present Reality Defender's submission to the ASVspoof5 challenge, highlighting a novel pretraining strategy which significantly improves generalizability while maintaining low computational cost during training. Our system SLIM learns the style-linguistics dependency embeddings from various types of bonafide speech using self-supervised contrastive learning. The learned embeddings help to discriminate spoof from bonafide speech by focusing on the relationship between the style and linguistics aspects. We evaluated our system on ASVspoof5, ASV2019, and In-the-wild. Our submission achieved minDCF of 0.1499 and EER of 5.5% on ASVspoof5 Track 1, and EER of 7.4% and 10.8% on ASV2019 and In-the-wild respectively.