mlsanityData to 14 Oct 2026

Hearing the Unspoken: Language Model Priors for Acoustic Adversarial Attacks

arXiv 2606.06833 ↗First posted 5 Jun 2026cs.LG cs.AI cs.CRcited by 0

Not on the map. Why?

Abstract (arXiv)

Automatic Speech Recognition (ASR) systems operating in real-time settings must process acoustic input under strict temporal constraints, where transcription decisions are inherently made on incomplete information. This causal constraint serves as an information bottleneck on attackers, significantly limiting attack performance. Our new Semantic Gambit attack overcomes this causal limitation by augmenting the adversary with predictive context derived from a Large Language Model in real-time. Our experiments show that this form of augmentation can elevate the corpus-level Word Error Rate to 35.6%-a three-fold increase over the current state-of-the-art streaming attack. Ultimately, this work reveals how common, low-latency LLM tooling can be exploited to systematically subvert real-time ASR pipelines.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 36 references that are in the map. First posted on arXiv 5 Jun 2026.

Builds on

29 of 36 references.

References · 29

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2022Robust Speech Recognition via Large-Scale Weak Supervisioncited by 4,057
  5. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  6. 2021HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Unitscited by 2,342
  7. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,420
  8. 2019Common Voice: A Massively-Multilingual Speech Corpuscited by 1,384
  9. 2021Perceiver: General Perception with Iterative Attentioncited by 1,104
  10. 2023Large Language Models are not Fair Evaluatorscited by 1,085
  11. 2015Listen, Attend and Spellcited by 543
  12. 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 491
17 more
  1. 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 451
  2. 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 274
  3. 2019Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognitioncited by 124
  4. 2018Adversarial Attacks Against Automatic Speech Recognition Systems via Psychoacoustic Hidingcited by 85
  5. 2018CommanderSong: A Systematic Approach for Practical Adversarial Voice Recognitioncited by 66
  6. 2020Towards Fast and Accurate Streaming End-to-End ASRcited by 47
  7. 2018Robust Audio Adversarial Example for a Physical Attackcited by 44
  8. 2019Universal Adversarial Perturbations for Speech Recognition Systemscited by 41
  9. 2021WaveGuard: Understanding and Mitigating Audio Adversarial Examplescited by 40
  10. 2019RawNet: Advanced end-to-end deep neural network using raw waveforms for text-independent speaker verificationcited by 33
  11. 2019Perceptual Based Adversarial Audio Attackscited by 13
  12. 2019Real-Time Adversarial Attackscited by 11
  13. 2019Universal Adversarial Audio Perturbationscited by 11
  14. 2021Real-Time Neural Voice Camouflagecited by 6
  15. 2024ALIF: Low-Cost Adversarial Audio Attacks on Black-Box Speech Platforms using Linguistic Featurescited by 5
  16. 2021Beyond Lp clipping: Equalization-based Psychoacoustic Attacks against ASRscited by 4
  17. 2022Watch What You Pretrain For: Targeted, Transferable Adversarial Examples on Self-Supervised Speech Recognition modelscited by 2

Built on by

Not cited yet.