mlsanityData to 7 Oct 2026

Natural Backdoor Attacks on Speech Recognition Models

arXiv 2607.15724 ↗First posted 17 Jul 2026cs.CR cs.LG cs.SDcited by 0

On the map: LLM safety and security → Audio jailbreaks and speech backdoors

Abstract (arXiv)

With the rapid development of deep learning, its vulnerability has gradually emerged in recent years. This work focuses on backdoor attacks on speech recognition systems. We adopt sounds that are ordinary in nature or in our daily life as triggers for natural backdoor attacks. We conduct experiments on two datasets and three models to validate the performance of natural backdoor attacks and explore the effects of poisoning rate, trigger duration and blend ratio on the performance of natural backdoor attacks. Our results show that natural backdoor attacks have a high attack success rate without compromising model performance on benign samples, even with short or low-amplitude triggers. It requires only 5% of poisoned samples to achieve a near 100% attack success rate. In addition, the backdoor will be automatically activated by the corresponding sound in nature, which is not easy to be detected and will bring severer harm.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Audio jailbreaks and speech backdoors, one of 12 directions in LLM safety and security. Builds on 17 of its 26 references that are in the map. First posted on arXiv 17 Jul 2026.

Builds on

17 of 26 references.

References · 17

  1. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  2. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  3. 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,471
  4. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  5. 2019Towards Federated Learning at Scale: System Designcited by 826
  6. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  7. 2018Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Surveycited by 459
  8. 2020Backdoor Learning: A Surveycited by 412
  9. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  10. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Triggercited by 203
  11. 2019Detecting AI Trojans Using Meta Neural Analysiscited by 192
  12. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfercited by 147
5 more
  1. 2020An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networkscited by 92
  2. 2020Backdoor Attack against Speaker Verificationcited by 62
  3. 2018A neural attention model for speech command recognitioncited by 44
  4. 2021Can You Hear It? Backdoor Attacks via Ultrasonic Triggerscited by 41
  5. 2019Adversarial Example Detection by Classification for Deep Speech Recognitioncited by 10

Built on by

Not cited yet.