Zero-Query Adversarial Attack on Black-box Automatic Speech Recognition Systems

arXiv:2406.19311v1 Announce Type: new
Abstract: In recent years, extensive research has been conducted on the vulnerability of ASR systems, revealing that black-box adversarial example attacks pose significant threats to real-world ASR systems. However, most existing black-box attacks rely on queries to the target ASRs, which is impractical when queries are not permitted. In this paper, we propose ZQ-Attack, a transfer-based adversarial attack on ASR systems in the zero-query black-box setting. Through a comprehensive review and categorization of modern ASR technologies, …

adversarial adversarial attack arxiv asr attack attacks automatic box cs.cr cs.sd eess.as queries query real recognition research speech speech recognition systems target threats vulnerability world