arXiv Machine Learning November 2, 2026 By Shengding Liu, Ashwin Innuganti, Qiben Yan

GhostPrefix: Synchronization-Free Universal Prefix Attacks for ASR Manipulation

Read the original on arXiv Machine Learning →

GhostPrefix is a new synchronization‑free universal adversarial prefix attack that learns a short audio segment to prepend to any live speech, enabling real‑time manipulation of automatic speech recognition (ASR) outputs without needing future utterance knowledge or temporal alignment. The method combines token‑level and representation‑level objectives to maintain stable effects across varying speech contexts and supports prompted, targeted, and untargeted attack modes. Evaluations on 13 open‑source ASR models show high success rates (up to 97.6% in digital settings) and significant impact on downstream AI‑assisted pipelines such as voice assistants and note‑taking tools.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv AI

Jun 8

Hearing the Unspoken: Language Model Priors for Acoustic Adversarial Attacks

arXiv:2606. 06833v1 Announce Type: cross Abstract: 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.

By Jiani Xie, Andrew C. Cullen, Paul Montague, Benjamin I. P. Rubinstein

llms multimodal benchmarks safety

arXiv AI

Sep 12

Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems

The paper introduces the Spectral Masking and Interpolation Attack (SMIA), a black‑box adversarial technique that subtly alters inaudible frequency regions of AI‑generated audio to fool voice authentication systems and their countermeasures. Experiments show SMIA achieves at least 82% success against combined verification and countermeasure systems, 97.5% against standalone speaker verification, and 100% against countermeasures, revealing a critical security gap. The authors argue that current static defenses are inadequate and call for dynamic, context‑aware defenses that can adapt to evolving threats.

By Kamel Kamel, Hridoy Sankar Dutta, Keshav Sood, Sunil Aryal

multimodal benchmarks safety

arXiv Machine Learning

Oct 21

You Are the Trigger: Speaker-Activated Backdoors

arXiv:2610.27210v1 Announce Type: new Abstract: Speech models are increasingly used as an interface to AI systems, from transcription tools to voice agents. Existing audio backdoors require an attack...

By Nata\v{s}a Kr\v{c}o, Euodia Dodd, Peter Romov, Yves-Alexandre de Montjoye

llms agents fine-tuning multimodal

arXiv AI

Jun 6

Beyond Waveform Robustness: Robust Feature-Vocoder Adversarial Attacks on Automatic Speech Recognition

arXiv:2606. 05678v1 Announce Type: cross Abstract: Automatic speech recognition (ASR) systems have become widely used for multilingual speech-to-text transcription.

By Yifan Liao, Zongmin Zhang, Zhen Sun, Yuhui Sun, Xinhu Zheng, Xinlei He

multimodal benchmarks safety

arXiv AI

Oct 15

Adaptive Compression and Targeted Perturbation: A Unified Framework for Generalized Audio Deepfake Detection

arXiv:2610.16049v1 Announce Type: cross Abstract: Existing audio deepfake detection (ADD) methods frequently exhibit limited generalization, as fixed bottleneck architectures fail to adapt to the het...

By Jianqiao Cui, Haoxian Zhu, Hanbiao Meng, Yanghao Zhang

multimodal benchmarks safety

arXiv AI

Aug 12

Never Stop Speaking: a Denial-of-Service Attack on End-to-End Speech Language Models

arXiv:2608. 10405v1 Announce Type: cross Abstract: Many studies have shown that specially crafted inputs can induce large language models (LLMs) to generate excessively long outputs, resulting in significant computational overhead and resource consumption.

By Shuozhe Cheng, Kunlan Xiang, Mingxuan Li, Ji Zhang, Dongxiao Liu, Wenbo Jiang

llms robotics multimodal safety

One email a morning, machine-written

Related stories

Hearing the Unspoken: Language Model Priors for Acoustic Adversarial Attacks

Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems

You Are the Trigger: Speaker-Activated Backdoors

Beyond Waveform Robustness: Robust Feature-Vocoder Adversarial Attacks on Automatic Speech Recognition

Adaptive Compression and Targeted Perturbation: A Unified Framework for Generalized Audio Deepfake Detection

Never Stop Speaking: a Denial-of-Service Attack on End-to-End Speech Language Models