Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks

Sound event detection and classification present significant challenges, particularly in noisy environments with multiple overlapping sources. This paper introduces an innovative architecture for multiple sound event detection and classification utilizing recurrent spiking neural networks (SNNs). Ou...

Full description

Saved in:
Bibliographic Details
Main Authors: Zahra Roozbehi, Ajit Narayanan, Mahsa Mohaghegh, Samaneh-Alsadat Saeedinia
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10973233/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850278990302937088
author Zahra Roozbehi
Ajit Narayanan
Mahsa Mohaghegh
Samaneh-Alsadat Saeedinia
author_facet Zahra Roozbehi
Ajit Narayanan
Mahsa Mohaghegh
Samaneh-Alsadat Saeedinia
author_sort Zahra Roozbehi
collection DOAJ
description Sound event detection and classification present significant challenges, particularly in noisy environments with multiple overlapping sources. This paper introduces an innovative architecture for multiple sound event detection and classification utilizing recurrent spiking neural networks (SNNs). Our method uniquely leverages temporal data to detect and classify multiple sound sources simultaneously, integrating the physical concept of signal power matching with neuronal output power and employing a binaural strategy to enhance detection accuracy in real-world scenarios. The architecture processes spatiotemporal data to dynamically update synaptic weights, enabling precise identification of sound event categories and their occurrences. Our simulations reveal substantial performance improvements, achieving the highest precision of 73% in classification tasks, including multilayer perceptrons (MLP), convolutional recurrent neural networks (CRNN), and recurrent neural networks (RNN). Statistical analysis indicates that these improvements are significant (p-value ¡ 0.05). These findings suggest practical applications in various fields such as surveillance, autonomous vehicles, and smart home systems, where robust sound event detection is critical.
format Article
id doaj-art-b13aae6628ec411ca9d70efcfdbc0ad5
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-b13aae6628ec411ca9d70efcfdbc0ad52025-08-20T01:49:15ZengIEEEIEEE Access2169-35362025-01-0113813128132510.1109/ACCESS.2025.356334610973233Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural NetworksZahra Roozbehi0https://orcid.org/0000-0003-1440-7015Ajit Narayanan1https://orcid.org/0000-0003-3075-7672Mahsa Mohaghegh2https://orcid.org/0000-0003-2228-8300Samaneh-Alsadat Saeedinia3https://orcid.org/0009-0000-3771-8073School of Electrical and Computer Engineering, Auckland University of Technology, Auckland, New ZealandSchool of Electrical and Computer Engineering, Auckland University of Technology, Auckland, New ZealandSchool of Electrical and Computer Engineering, Auckland University of Technology, Auckland, New ZealandDepartment of Electrical and Computer Engineering, Iran University of Science and Technology, Tehran, IranSound event detection and classification present significant challenges, particularly in noisy environments with multiple overlapping sources. This paper introduces an innovative architecture for multiple sound event detection and classification utilizing recurrent spiking neural networks (SNNs). Our method uniquely leverages temporal data to detect and classify multiple sound sources simultaneously, integrating the physical concept of signal power matching with neuronal output power and employing a binaural strategy to enhance detection accuracy in real-world scenarios. The architecture processes spatiotemporal data to dynamically update synaptic weights, enabling precise identification of sound event categories and their occurrences. Our simulations reveal substantial performance improvements, achieving the highest precision of 73% in classification tasks, including multilayer perceptrons (MLP), convolutional recurrent neural networks (CRNN), and recurrent neural networks (RNN). Statistical analysis indicates that these improvements are significant (p-value ¡ 0.05). These findings suggest practical applications in various fields such as surveillance, autonomous vehicles, and smart home systems, where robust sound event detection is critical.https://ieeexplore.ieee.org/document/10973233/Sound classificationspiking neural networkTempotronaudio signal powerdetectionmultiple sound sources
spellingShingle Zahra Roozbehi
Ajit Narayanan
Mahsa Mohaghegh
Samaneh-Alsadat Saeedinia
Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks
IEEE Access
Sound classification
spiking neural network
Tempotron
audio signal power
detection
multiple sound sources
title Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks
title_full Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks
title_fullStr Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks
title_full_unstemmed Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks
title_short Enhanced Multiple Sound Event Detection and Classification Using Physical Signal Properties in Recurrent Spiking Neural Networks
title_sort enhanced multiple sound event detection and classification using physical signal properties in recurrent spiking neural networks
topic Sound classification
spiking neural network
Tempotron
audio signal power
detection
multiple sound sources
url https://ieeexplore.ieee.org/document/10973233/
work_keys_str_mv AT zahraroozbehi enhancedmultiplesoundeventdetectionandclassificationusingphysicalsignalpropertiesinrecurrentspikingneuralnetworks
AT ajitnarayanan enhancedmultiplesoundeventdetectionandclassificationusingphysicalsignalpropertiesinrecurrentspikingneuralnetworks
AT mahsamohaghegh enhancedmultiplesoundeventdetectionandclassificationusingphysicalsignalpropertiesinrecurrentspikingneuralnetworks
AT samanehalsadatsaeedinia enhancedmultiplesoundeventdetectionandclassificationusingphysicalsignalpropertiesinrecurrentspikingneuralnetworks