Abstract
Digital audio recording is the main evidence used in the field of judicial forensics. Recently, a number of digital audio forensic techniques have been developed and the audio source identification (ASI) is one of the most active research topics. Most of existing ASI works mainly focus on improving the performance of detection accuracy and robustness. Little consideration has been given to ASI anti-forensics, which aims at attacking the forensic techniques. To expose the weaknesses of these source identification methods, we propose an anti-forensic framework based on generative adversarial network (GAN) to falsify the source information of an audio by adding specific disturbance. The experimental results show that the falsified audio can deceive the forensic methods effectively, and can even control their conclusions. Three state-of-art ASI methods have been evaluated as the attacking targets. For the confusing attack, the proposed method can significantly reduce their detection accuracies from about 97% to less than 5%. For the misleading attack, a misleading rate about 81.32% has been achieved while ensuring the perceptual quality of the anti-forensic audio.
Highlights
Audio source identification (ASI) is an important part of digital audio forensics, which focus on identifying the source of the target audio from a candidate list of recording devices
EXPERIMENTAL RESULTS we evaluate our proposed anti-forensic methods by attacking the current state-of-the-art ASI techniques
In this paper, we present an anti-forensic method to against the audio source identification technique, which based on a generative adversarial network
Summary
Audio source identification (ASI) is an important part of digital audio forensics, which focus on identifying the source of the target audio from a candidate list of recording devices. To construct our anti-forensic attack, we design and train our GAN architecture, and we use the trained generator to falsify the original specific device audio. The overall goal of our anti-forensic attack is to falsify the forensic traces left by a recording device, so that the attacked audio can spoof the source identification classifier. C. LOSS FUNCTION To achieve the anti-forensic purpose, the generator G should be able to automatically learn how to falsify an audio to deceive the classifier successful while introducing a minimal disturbance. In order to realize an effective confusing attack, we need to optimize the generator using the classification loss Lc1 and Lc2, which are designed to measure the distance between the output of C in classifying the falsified audio x’ with its original category y. ACC (Accuracy) and SAR (Successful Attack Rate) are adopted to demonstrate the attack performance and Perceptual evaluation of speech quality (PESQ) [21] is to evaluate the perceptual quality of the falsified audio
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
