mlsanityData to 14 Oct 2026

Decoding Deception: Understanding Automatic Speech Recognition Vulnerabilities in Evasion and Poisoning Attacks

arXiv 2509.22060 ↗First posted 26 Sep 2025cs.SD cs.AI cs.CRcited by 0

Not on the map. Why?

Abstract (arXiv)

Recent studies have demonstrated the vulnerability of Automatic Speech Recognition systems to adversarial examples, which can deceive these systems into misinterpreting input speech commands. While previous research has primarily focused on white-box attacks with constrained optimizations, and transferability based black-box attacks against commercial Automatic Speech Recognition devices, this paper explores cost efficient white-box attack and non transferability black-box adversarial attacks on Automatic Speech Recognition systems, drawing insights from approaches such as Fast Gradient Sign Method and Zeroth-Order Optimization. Further, the novelty of the paper includes how poisoning attack can degrade the performances of state-of-the-art models leading to misinterpretation of audio signals. Through experimentation and analysis, we illustrate how hybrid models can generate subtle yet impactful adversarial examples with very little perturbation having Signal Noise Ratio of 35dB that can be generated within a minute. These vulnerabilities of state-of-the-art open source model have practical security implications, and emphasize the need for adversarial security.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 20 of its 35 references that are in the map. First posted on arXiv 26 Sep 2025.

Builds on

20 of 35 references.

References · 20

  1. 2022Generative Adversarial Networkscited by 5,627
  2. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  3. 2016Adversarial examples in the physical worldcited by 1,769
  4. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
  5. 2017Evasion Attacks against Machine Learning at Test Timecited by 689
  6. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 666
  7. 2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 414
  8. 2019HopSkipJumpAttack: A Query-Efficient Decision-Based Attackcited by 333
  9. 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 274
  10. 2019Adversarial Attacks and Defenses in Images, Graphs and Text: A Reviewcited by 237
  11. 2017Towards Poisoning of Deep Learning Algorithms with Back-gradient Optimizationcited by 197
  12. 2018Exploring Connections Between Active Learning and Model Extractioncited by 85
8 more
  1. 2017DolphinAtack: Inaudible Voice Commandscited by 69
  2. 2018CommanderSong: A Systematic Approach for Practical Adversarial Voice Recognitioncited by 66
  3. 2018Cache Telepathy: Leveraging Shared Resource Attacks to Learn DNN Architecturescited by 57
  4. 2020SoK: The Faults in our ASRs: An Overview of Attacks against Automatic Speech Recognition and Speaker Identification Systemscited by 46
  5. 2019A Survey of Black-Box Adversarial Attacks on Computer Vision Modelscited by 37
  6. 2017Is Deep Learning Safe for Robot Vision? Adversarial Examples against the iCub Humanoidcited by 19
  7. 2018Understanding and Mitigating the Security Risks of Voice-Controlled Third-Party Skills on Amazon Alexa and Google Homecited by 3
  8. 2020Preliminary Study of a Google Home Minicited by 1

Built on by

Not cited yet.