ABSTRACT
DolphinAttack is an attack model aiming at audio frequency identification vulnerabilities in voice control system microphones. It modulates voice command on the ultrasonic carrier to achieve inaudibility, and the modulated low frequency audio command can be successfully demodulated, recorded, and interpreted by the speech recognition system. The DolphinAttack threatens the property right and privacy of smart device users. In the paper, we propose a security reinforcement method for voice control system based on disconnected loop authentication that combines software and hardware to prevent DolphinAttack. The security enhancement solution that we proposed based on secondary authentication also can mitigate some unknown vulnerabilities. When the attacker or user wakes up the speech recognition system and performs command control, it will force the voice control system at a high security level to perform secondary authentication. If the command passes the secondary authentication, it enters the execution state. Otherwise, the command is discarded and the report is sent to the user. This kind of security enhancement avoids the privacy leakage and property damage caused by the DolphinAttack at some degree.
- Your home at your command. [Online]. Available URL: https://www.apple.com/ios/home/ (2017-12-14).Google Scholar
- Ruohan Meng, Steven G. Rice, Jin Wang and Xingming Sun. A fusion steganographic algorithm based on faster R-CNN[J]. Computers, Materials & Continua, 2018, 55(1): 1--16.Google Scholar
- Nicholas Carlini, Pratyush Mishra, Tavish Vaidya, Yuankai Zhang, Micah Sherr, Clay Shields, David Wagner, and Wenchao Zhou. Hidden voice commands. In Proceedings of the USENIX Security Symposium 2016. Google ScholarDigital Library
- Yi Cao, Zhili Zhou, Xingming Sun and Chongzhi Gao. Coverless information hiding based on the molecular structure images of material[J]. Computers, Materials & Continua, 2018, 54 (2): 197--207.Google Scholar
- Guoming Zhang, Chen Yan, Xiaoyu Ji, Tianchen Zhang, Taimin Zhang, and Wenyuan Xu. DolphinAttack: Inaudible Voice Commands. In Proceedings of the 2017 ACM SIGSAC Conf. on Computer and Communications Security (CCS) Google ScholarDigital Library
- Lea Schönherr, Katharina Kohls, Steffen Zeiler, Thorsten Holz, and Dorothea Kolossa. Adversarial attacks against automatic speech recognition systems via psychoacoustic hiding. arXiv preprint arXiv:1808.05665, 2018.Google Scholar
- Microphone. [Online]. Available URL: https://en.wikipedia.org/wiki/Microphone (2018-12-09)Google Scholar
- Dibya Mukhopadhyay, Maliheh Shirvanian, and Nitesh Saxena. All your voices are belong to us: Stealing voices to fool humans and machines. In Proceedings of the European Symposium on Research in Computer Security. Springer, 2015.Google ScholarCross Ref
- CommanderSong. [Online]. Available URL: https://mp.weixin.qq.com/s/VnhSCFVeQwHuYRFqL5JP4g (2020-12-01)Google Scholar
Index Terms
- A defense scheme of voice control system against DolphinAttack
Recommendations
DolphinAttack: Inaudible Voice Commands
CCS '17: Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications SecuritySpeech recognition (SR) systems such as Siri or Google Now have become an increasingly popular human-computer interaction method, and have turned various systems into voice controllable systems (VCS). Prior work on attacking VCS shows that the hidden ...
Toward an automatic speech recognition system for amazigh-tarifit language
This work aims at contributing to the Amazigh language Automatic Speech Recognition (ASR). We have studied and realized an automatic speech recognition system, using an environment totally based on the Amazigh-Tarifit language. In this framework, we ...
A DDOS defence scheme based on two-stage traffic flow control
In distributed denial of service (DDoS) attack, the detection techniques have existed for a relatively longer period of time than defence mechanisms, researchers have categorised almost all the existing and expected forthcoming attacks. However, ...
Comments