Skip to content
conftrace_
2025 Interspeech INTERSPEECH 2025

Replay Attacks Against Audio Deepfake Detection

Abstract

We show how replay attacks undermine audio deepfake detection: By playing and re-recording deepfake audio through various speakers and microphones, we make spoofed samples appear authentic to the detection model.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Speech & Audio

Authors

Nicolas Müller, Piotr Kawa, Wei-Herng Choong, Adriana Stan, Aditya Tirumala Bukkapatnam, Karla Pizzi, Alexander Wagner, Philip Sperl

Topics

Speech & Audio > Analysis > Speech Analysis Artificial Intelligence > Core AI > Adversarial Learning Artificial Intelligence > Core AI > Security

Keywords

audio deepfake detection re-recording attack replay attack spoofed sample
Download PDF

More from Interspeech 2025

All Interspeech 2025 papers →
Bona fide Cross Testing Reveals Weak Spot in Audio Deepfake Detection Systems
BiCrossMamba-ST: Speech Deepfake Detection with Bidirectional Mamba Spectro-Temporal Cross-Attention
Few-Shot Speech Deepfake Detection Adaptation with Gaussian Processes
Enhancing Audio Deepfake Detection by Improving Representation Similarity of Bonafide Speech
Generalizable Audio Deepfake Detection via Hierarchical Structure Learning and Feature Whitening in Poincaré sphere
VoiceNet: Multilingual On-Device Phoneme-To-Audio Alignment