Stroud, S, Jones, K, Edwards, G, Robinson, C, Chandler-Crnigoj, S and Ellis, D (2024) Enhancing Environmental Sound Recognition in Digital Simulations: A Novel Approach to Beamforming and Signal Identification. In: Proceedings of the International Conference on Computer Systems and Technologies 2024 . pp. 167-172. (CompSysTech '24: International Conference on Computer Systems and Technologies 2024, 14th Jun - 15th Jun 2024, Ruse, Bulgaria).

Preview

Text Enhancing Environmental Sound Recognition in Digital Simulations A Novel Approach to Beamforming and Signal Identification.pdf - Accepted Version Download (1MB) | Preview

Abstract

This paper advances the field of environmental sound recognition, presenting a refined approach to beamforming and noise identification through digital simulations of realistic environments at Liverpool John Moores University. Amidst the growing demand for precise audio detection techniques in cluttered acoustic environments, our research introduces a method for identifying and highlighting specific sound signals. We use an advanced time-delay beamforming algorithm to achieve strategic audio zooming, addressing topical issues in urban surveillance and forensic sound examination for potential analysis in criminal cases. Our methodology is rooted in deploying a carefully configured array of virtual omnidirectional microphones, crucial for collecting real-world audio signals. Our technique's core lies in applying our advanced algorithm to the captured sound data, thoroughly assessing our system's capability to identify and isolate targeted sound sources. Our investigation further measures the robustness of our system to microphone failure, which continues to function even when microphones completely fail, highlighting its reliability, even when operating under compromised conditions. Through simulations that capture actual acoustic environments, our experiments reveal the algorithm's proficiency in coping with both sound reflections and reverberations, critical elements in authentically reproducing real-world scenarios. Finally, this study explores the extended applicability of our research findings, considering their potential impact across various sectors, including environmental surveillance, animal conservation, broadcasting, and sound engineering. Our work offers a forward-thinking strategy for environmental sound recognition within a digital simulation framework. It paves the way for practical applications that stand to gain from improved sound separation and analysis techniques. Our contributions engage with the broader dialogue on the evolution of surveillance technology, providing valuable perspectives that could influence the future of audio research.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	© S. Stroud | ACM 2024. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in ACM Digital Library, http://doi.org/10.1145/3674912.3674940
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) T Technology > TK Electrical engineering. Electronics. Nuclear engineering
Divisions:	Engineering
Publisher:	ACM
SWORD Depositor:	A Symplectic
Date Deposited:	14 Aug 2024 08:32
Last Modified:	14 Aug 2024 08:33
DOI or ID number:	10.1145/3674912.3674940
URI:	https://researchonline.ljmu.ac.uk/id/eprint/23909

View Item