Search papers, labs, and topics across Lattice.
This paper introduces PV-VASM, a probabilistic framework for verifying the robustness of voice anti-spoofing models (VASMs) against various spoofing attacks. PV-VASM estimates the probability of misclassification under text-to-speech (TTS), voice cloning (VC), and parametric signal transformations, offering a model-agnostic approach to robustness verification. Experiments across diverse settings validate the method's effectiveness, demonstrating its ability to generalize to unseen speech synthesis techniques and input perturbations while providing a theoretical upper bound on the error probability.
Uncover the hidden vulnerabilities of your voice anti-spoofing model with a new tool that quantifies the probability of failure against unseen speech synthesis attacks.
Recent advances in generative models have amplified the risk of malicious misuse of speech synthesis technologies, enabling adversaries to impersonate target speakers and access sensitive resources. Although speech deepfake detection has progressed rapidly, most existing countermeasures lack formal robustness guarantees or fail to generalize to unseen generation techniques. We propose PV-VASM, a probabilistic framework for verifying the robustness of voice anti-spoofing models (VASMs). PV-VASM estimates the probability of misclassification under text-to-speech (TTS), voice cloning (VC), and parametric signal transformations. The approach is model-agnostic and enables robustness verification against unseen speech synthesis techniques and input perturbations. We derive a theoretical upper bound on the error probability and validate the method across diverse experimental settings, demonstrating its effectiveness as a practical robustness verification tool.