Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognition
Yao Qin · Nicholas Carlini · Garrison Cottrell · Ian Goodfellow · Colin Raffel

Wed Jun 12th 11:35 -- 11:40 AM @ Grand Ballroom

Adversarial examples are inputs to machine learning models designed by an adversary to cause an incorrect output. So far, adversarial examples have been studied most extensively in the image domain. In this domain, adversarial examples can be constructed by imperceptibly modifying images to cause misclassification, and are practical in the physical world. In contrast, current targeted adversarial examples on speech recognition systems have neither of these properties: humans can easily identify the adversarial perturbations, and they are not effective when played over-the-air. This paper makes progress on both of these fronts. First, we develop effectively imperceptible audio adversarial examples (verified through a human study) by leveraging the psychoacoustic principle of auditory masking, while retaining 100% targeted success rate on arbitrary full-sentence targets. Then, we make progress towards physical-world audio adversarial examples by constructing perturbations which remain effective even after applying highly-realistic simulated environmental distortions.

Author Information

Yao Qin (University of California, San Diego)
Nicholas Carlini (Google)
Garrison Cottrell (University of California, San Diego)
Ian Goodfellow (Google Brain)
Colin Raffel (Google)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors