In the same section
Trends and Challenges in Audio Forensics
Chaired by Sara Mandelli and Paolo Bestagini, both from Politecnico di Milano, Italy, and Matthew Stamm, Drexel University, Philadelphia, USA.
In the last few years, we have witnessed a series of technological advancements related to multimedia content acquisition and editing. On one hand, this means that it is possible to exploit new kinds of data for forensics analysis (e.g., consider images and videos shot with multi-camera smartphones). On the other hand, this means that an attacker has an improved arsenal of operations that can be maliciously applied to forge multimedia objects (e.g., consider the possibilities offered by deepfake solutions). It is therefore important for the multimedia forensics community to keep the pace and develop modern forensic solutions.
Unfortunately, great part of the developed modern forensic solutions focuses on the analysis of still images and videos. Audio forensic solutions are often overlooked. However, the new technological trends should also be considered in the audio realm. On one hand, modern smartphones equipped with multiple microphones can be a useful asset for forensic analysts willing to monitor an acoustic scene or environment. On the other hand, audio deepfake and voice conversion techniques enable to swap the vocal identity of a speaker, thus calling for the development of new detectors.
This special session wants to bring the forensic community attention to the recently overlooked audio forensic field, by showing actual trends and challenges. The goal is to discuss modern audio forensic techniques to showcase both the potential and the risks brought by recent technological advancements in the audio realm, from acoustic scene sensing to device identification and synthetic audio detection. This will hopefully raise the awareness of the research community on modern audio forensics.
Important note: all papers (for special and regular sessions) must be submitted on Conference Management Toolkit (CMT) following this link.For this special session in particular, authors must select the track "SPECIAL SESSION" on CMT website during the submission.
Contact: If you have any questions feel free to contact us at wifs2022@utt.fr.