MAIA: An Inpainting-Based Approach for Music Adversarial Attacks
By: Yuxuan Liu , Peihong Zhang , Rui Sang and more
Potential Business Impact:
Tricks music AI into hearing wrong notes.
Music adversarial attacks have garnered significant interest in the field of Music Information Retrieval (MIR). In this paper, we present Music Adversarial Inpainting Attack (MAIA), a novel adversarial attack framework that supports both white-box and black-box attack scenarios. MAIA begins with an importance analysis to identify critical audio segments, which are then targeted for modification. Utilizing generative inpainting models, these segments are reconstructed with guidance from the output of the attacked model, ensuring subtle and effective adversarial perturbations. We evaluate MAIA on multiple MIR tasks, demonstrating high attack success rates in both white-box and black-box settings while maintaining minimal perceptual distortion. Additionally, subjective listening tests confirm the high audio fidelity of the adversarial samples. Our findings highlight vulnerabilities in current MIR systems and emphasize the need for more robust and secure models.
Similar Papers
Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
Sound
Makes fake voices trick voice locks.
AdvPaint: Protecting Images from Inpainting Manipulation via Adversarial Attention Disruption
CV and Pattern Recognition
Stops fake pictures from replacing parts of photos.
Membership and Dataset Inference Attacks on Large Audio Generative Models
Machine Learning (CS)
Finds if artists' music trained AI.