EnvSDD: Benchmarking Environmental Sound Deepfake Detection

TL;DR

This paper introduces EnvSDD, a large-scale dataset for environmental sound deepfake detection, and proposes a detection system that outperforms existing methods, addressing the unique challenges of environmental sounds.

Contribution

The paper provides the first large-scale dataset for environmental sound deepfake detection and develops a novel detection system tailored for this domain.

Findings

The proposed system outperforms state-of-the-art methods.

EnvSDD includes diverse conditions for robust evaluation.

The dataset covers extensive real and fake environmental sounds.

Abstract

Audio generation systems now create very realistic soundscapes that can enhance media production, but also pose potential risks. Several studies have examined deepfakes in speech or singing voice. However, environmental sounds have different characteristics, which may make methods for detecting speech and singing deepfakes less effective for real-world sounds. In addition, existing datasets for environmental sound deepfake detection are limited in scale and audio types. To address this gap, we introduce EnvSDD, the first large-scale curated dataset designed for this task, consisting of 45.25 hours of real and 316.74 hours of fake audio. The test set includes diverse conditions to evaluate the generalizability, such as unseen generation models and unseen datasets. We also propose an audio deepfake detection system, based on a pre-trained audio foundation model. Results on EnvSDD show that our proposed system outperforms the state-of-the-art systems from speech and singing domains.