Interpretable Spectrum Transformation Attacks to Speaker Recognition Systems

• Published in: IEEE/ACM transactions on audio, speech, and language processing Vol. 32; pp. 1531 - 1545
• Main Authors: Yao, Jiadi, Luo, Hong, Qi, Jun, Zhang, Xiao-Lei
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: adversarial examples; adversarial transferability; Black boxes; black-box attacks; Closed box; Data models; Discrete cosine transform; Frequencies; Frequency domain analysis; Glass box; Optimization; Perturbation methods; Speaker recognition; Speech recognition; Time-frequency analysis

The success of adversarial attacks on speaker recognition is mainly in white-box scenarios. When applying the adversarial voices that are generated by attacking white-box surrogate models to black-box victim models, i.e. transfer-based black-box attacks, the transferability of the adversarial voices...