Towards Robust Speech Models: Mitigating Backdoor Attacks via Audio Signal Enhancement and Fine-Pruning Techniques
The widespread adoption of deep neural networks (DNNs) in speech recognition has introduced significant security vulnerabilities, particularly from backdoor attacks. These attacks allow adversaries to manipulate system behavior through hidden triggers while maintaining normal operation on clean inpu...
Saved in:
| Main Authors: | Heyan Sun, Qi Zhong, Minfeng Qi, Uno Fang, Guoyi Shi, Sanshuai Cui |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/6/984 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Similar Items
-
Backdoor Defence for Voice Print Recognition Model Based on Speech Enhancement and Weight Pruning
by: Jiawei Zhu, et al.
Published: (2022-01-01) -
A4FL: Federated Adversarial Defense via Adversarial Training and Pruning Against Backdoor Attack
by: Saeed-Uz-Zaman, et al.
Published: (2025-01-01) -
Efficient Method for Robust Backdoor Detection and Removal in Feature Space Using Clean Data
by: Donik Vrsnak, et al.
Published: (2025-01-01) -
A survey of backdoor attacks and defences: From deep neural networks to large language models
by: Ling-Xin Jin, et al.
Published: (2025-09-01) -
Backdoor defense method in federated learning based on contrastive training
by: Jiale ZHANG, et al.
Published: (2024-03-01)