SoK - The Faults in Our ASRs - An Overview of Attacks Against Automatic Speech Recognition and Speaker Identification Systems

This course provides an overview of attacks against Automatic Speech Recognition (ASR) and Speaker Identification Systems. The learning outcomes include understanding different types of attacks, metrics for evaluating attacks, and the transferability of attacks. The course teaches skills related to identifying vulnerabilities in ASR systems and conducting experiments to test attack methods. The teaching method involves presenting information on attack techniques and their implications. The intended audience for this course includes cybersecurity professionals, researchers, and individuals interested in speech recognition technology.