A Comparative Study on Proactive and Passive Detection of Deepfake Speech
By: Chia-Hua Wu , Wanying Ge , Xin Wang and more
Potential Business Impact:
Tests fake voices to find the real ones.
Solutions for defending against deepfake speech fall into two categories: proactive watermarking models and passive conventional deepfake detectors. While both address common threats, their differences in training, optimization, and evaluation prevent a unified protocol for joint evaluation and selecting the best solutions for different cases. This work proposes a framework to evaluate both model types in deepfake speech detection. To ensure fair comparison and minimize discrepancies, all models were trained and tested on common datasets, with performance evaluated using a shared metric. We also analyze their robustness against various adversarial attacks, showing that different models exhibit distinct vulnerabilities to different speech attribute distortions. Our training and evaluation code is available at Github.
Similar Papers
Measuring the Robustness of Audio Deepfake Detectors
Cryptography and Security
Finds fake voices even when they are noisy.
Big Brother is Watching: Proactive Deepfake Detection via Learnable Hidden Face
CV and Pattern Recognition
Hides a secret face to catch fake videos.
Comparative Analysis of Deepfake Detection Models: New Approaches and Perspectives
CV and Pattern Recognition
Finds fake videos to stop lies.