An Intervention-Based Framework for Shortcut Diagnosis in Spoofing Countermeasures
Provides a systematic tool for diagnosing shortcut dependencies in spoofing countermeasures, addressing a critical gap for practitioners deploying audio forensic systems.
Deepfake audio detection models often rely on dataset-specific shortcuts, reducing real-world reliability. The proposed intervention-based framework identifies non-speech intervals as a dominant shortcut, causing the largest performance shifts in XLS-R-300M and RawGAT-ST models across ASVspoof datasets.
While deepfake audio detection systems achieve high performance in controlled benchmarks, their reliability often diminishes in the wild. Prior work shows that dataset-specific artifacts contribute to this gap. Yet, systematic tools to identify which acoustic properties a model exploits as shortcuts remain limited. We propose an intervention-based diagnostic framework, grounded in a directed graphical model, that formally distinguishes confound-driven shortcut dependencies from legitimate domain shift. We operationalise this through controlled acoustic perturbations targeting non-speech structure, spectral content, and signal energy, complemented by corpus-level distributional analysis. Evaluating XLS-R-300M with RawGAT-ST across ASVspoof challenges datasets, we quantify model sensitivity to specific intervention types. Results reveal that non-speech interventions produce the largest performance shifts, confirming non-speech intervals as a dominant shortcut.