Browsing by Author "Robinson, C"

Now showing 1 - 2 of 2

Open Access
Advancing Audio Surveillance in Simulated Environments: Real-World Soundscapes and Targeted Noise Detection through Enhanced Beamforming Techniques
(SLIIT, Faculty of Engineering, 2025-01) Stroud, S; Jones, K; Edwards, G; Robinson, C; Chandler-Crnigoj, S; Ellis, D
This paper introduces an innovative beamforming approach designed for audio surveillance, executed through a virtual simulation of a real-world environment based at Liverpool John Moores University. Our research is driven by the increasing requirement for sophisticated audio analysis methods to isolate and enhance specific sounds within noisy environments for forensic analysis, for example, in criminal court cases. By leveraging a time-delay beamforming algorithm, our work offers a novel solution to discern and amplify targeted noises amidst complex soundscapes, a challenge commonly encountered in urban surveillance and forensic audio analysis. Our approach's foundation lies in utilising a carefully arranged, robust array of omnidirectional microphones, which are instrumental in capturing a wide range of real-world sound signals. The core of our methodology involves processing captured sounds using the proposed algorithm, followed by evaluating the system's effectiveness in capturing the desired localised audio sources. This paper explores the system's resilience against microphone array degradation, showcasing its robustness in scenarios of partial system functionality. The experiments, grounded in the simulation of real-world acoustic environments, demonstrate the algorithm's adeptness at managing sound reflections and reverberation, critical factors in the realistic replication of urban soundscapes. It also considers the broader implications of our findings, exploring the potential for adopting this technology in various domains beyond law enforcement, including broadcast solutions, advanced audio engineering applications, and animal conservation in the wild. In conclusion, this research showcases a creative approach to audio surveillance and opens the door to numerous applications that can benefit from enhanced methods of audio isolation and analysis. Our findings contribute to the ongoing discourse on developing advanced surveillance technologies, offering insights that could help shape the future of audio processing and analysis.
Embargo
Evaluating the Threshold of Authenticity in Deepfake Audio and Its Implications Within Criminal Justice
(SLIIT, Faculty of Engineering, 2024-10) Rodgers, J; Jones, K.O; Robinson, C; Chandler-Crnigoj, S; Burrell, H; McColl, S
Deepfake technology has come a long way in recent years and the world has already seen cases where it has been used maliciously. After a deepfake of UK independent financial advisor and poverty champion Martin Lewis was released on social media, a theory has been proposed where the deepfake target is accompanied by additional media to increase the authenticity of the file, for instance, ambient noise or processing to match how the deepfake would sound if it was recorded from a specific device such as a cellular/mobile phone. Focussing on deepfake audio, a critical listening experiment was conducted where participants were asked to identify the deepfake audio file from a set of three, across a number of sets of three files. A number of audio files were created using real voices with additional sounds added, volunteers recording their voice which is then put through a deepfake generation system, and voices taken from publicly available podcasts which were also applied to the deepfake software – the latter set mimics using web accessible voice recordings of prominent or famous people, such as the Prime Minister of the UK. The results show participants were able to successfully detect one third of the deepfake audio files presented, however they also incorrectly marked another one third of the real files as deepfakes whilst the remaining third were missed. Results also showed no definitive confirmation that audio and/or forensic professionals had any greater ability to successfully detect deepfake audio files when compared to others. The false positive result may also reinforce the scepticism and lack of trust created by what is known as “Liar’s Dividend”. The paper details how the files were created, the testing methodology, and the experimental results. Furthermore, a discussion on the future directions of research and the effects that deepfakes may have on the criminal justice system is presented.