mlsanityData to 7 Oct 2026

SpeechGuard: Online Defense against Backdoor Attacks on Speech Recognition Models

arXiv 2607.15697 ↗First posted 17 Jul 2026cs.SD cs.CR cs.LGcited by 0

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

Abstract (arXiv)

Backdoor attacks pose a critical threat to neural network models, allowing attackers to implant a backdoor during the training phase by manipulating a small portion of the training data. In security-sensitive applications such as voice interaction for autonomous driving, the presence of backdoor attacks introduces substantial security risks. This study focuses on implementing backdoor defense measures for speech recognition models in run-time, taking into account the characteristics of audio signals. We propose SpeechGuard, the first online backdoor defense pipeline designed to identify and purify poisoned audio samples. Specifically, we improve STRIP method to perform adaptive perturbation injection to detect and filter poisoned samples, named as S-STRIP. More importantly, we further consider the purification of poisoned samples. We utilize time-frequency (T-F) masking to suppress the expression of trigger signals and autonomously generate masks based on an autoencoder. The two-stage processing prevents the backdoor in the model from being triggered, and even input speech carrying triggers can be accurately predicted. Extensive experimental demonstrate that SpeechGuard can accurately filter out poisoned samples. Through purification, it can significantly mitigate the backdoor threat while maintaining a certain prediction accuracy.

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 15 of its 24 references that are in the map. First posted on arXiv 17 Jul 2026.

Builds on

15 of 24 references.

References · 15

  1. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  2. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  3. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  4. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
  5. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 437
  6. 2020Backdoor Learning: A Surveycited by 412
  7. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  8. 2021Anti-Backdoor Learning: Training Clean Models on Poisoned Datacited by 258
  9. 2020Deep Feature Space Trojan Attack of Neural Networks by Controlled Detoxificationcited by 93
  10. 2020An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networkscited by 92
  11. 2020Backdoor Attack against Speaker Verificationcited by 62
  12. 2018A neural attention model for speech command recognitioncited by 44
3 more
  1. 2021Data Poisoning Attacks to Deep Learning Based Recommender Systemscited by 41
  2. 2021Can You Hear It? Backdoor Attacks via Ultrasonic Triggerscited by 41
  3. 2019Adversarial Example Detection by Classification for Deep Speech Recognitioncited by 10

Built on by

Not cited yet.