Adversarial Attacks On Audio Deepfake Detection: A Benchmark And Comparative Study

Abstract

The widespread use of generative AI has shown remarkable success in producing highly realistic deepfakes, posing a serious threat to various voice biometric applications, including speaker verification, voice biometrics, audio conferencing, and criminal investigations. To counteract this, several state-of-the-art (SoTA) audio deepfake detection (ADD) methods have been proposed to identify generative AI signatures to distinguish between real and deepfake audio. However, the effectiveness of these methods is severely undermined by anti-forensic (AF) attacks that conceal generative signatures. These AF attacks span a wide range of techniques, including statistical modifications (e.g., pitch shifting, filtering, noise addition, and quantization) and optimization-based attacks (e.g., FGSM, PGD, C \& W, and DeepFool). In this paper, we investigate the SoTA ADD methods and provide a comparative analysis to highlight their effectiveness in exposing deepfake signatures, as well as their vulnera

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyuddin2025adversarial

Related papers

• The Codecfake Dataset And Countermeasures For The Universally Detection Of Deepfake Audio (2024)10.97
• Benchmarking Audio Deepfake Detection Robustness In Real-world Communication Scenarios (2025)5.24
• Securing Voice Biometrics: One-shot Learning Approach For Audio Deepfake Detection (2023)9.03
• AUDETER: A Large-scale Dataset For Deepfake Audio Detection In Open Worlds (2025)0.00
• Towards Robust Audio Deepfake Detection: A Evolving Benchmark For Continual Learning (2024)0.00
• Pitch Imperfect: Detecting Audio Deepfakes Through Acoustic Prosodic Analysis (2025)0.00
• MFAAN: Unveiling Audio Deepfakes With A Multi-feature Authenticity Network (2023)7.81
• A Survey On Speech Deepfake Detection (2024)12.10