Enhanced audio classification with quantum-inspired neural layers and exponentially weighted attention fusion
An International Journal of Optimization and Control: Theories & Applications (IJOCTA) · 29.04.2026 · 16 (3) · 774
DOI: 10.36922/ijocta026050021 ↗Kurz gefasst
Fachartikel in der Zeitschrift An International Journal of Optimization and Control: Theories & Applications (IJOCTA), online erschienen am 29.04.2026. Verfasst von Zahraa Tarek und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Snoring is a common sleep problem and may indicate underlying health conditions, such as obstructive sleep apnea. Accurate detection and classification of snoring sounds are critical for early diagnosis and effective treatment. Traditional snoring detection methods rely on manual analysis or rule-based systems, which are often time-consuming and error-prone. With advances in deep learning and quantum-inspired computing, an opportunity has emerged to develop more powerful and intelligent solutions to automate snoring classification. In this paper, we propose a quantum-inspired attention fusion network for snoring detection, using Mel-Frequency Cepstral Coefficients as the primary representation of acoustic features. The network combines quantum-inspired layers with an attention mechanism to enhance feature learning and improve classification accuracy, offering a novel approach to audio analysis in the healthcare domain. The quantum-inspired layer simulates quantum gates using dense neural layers, enabling the model to detect complex patterns in audio data, while an attention mechanism dynamically weights the most important features to improve classification. The model achieved outstanding performance with an accuracy of 0.997, a prediction precision of 0.997, a recall of 0.997, and an average F1-score of 0.997, demonstrating its ability to generalize well and classify snoring sounds with near-perfect accuracy.
Zitieren
APA
Tarek, Z., & Hasan, E. (2026). Enhanced audio classification with quantum-inspired neural layers and exponentially weighted attention fusion. An International Journal of Optimization and Control: Theories & Applications (IJOCTA), 16(3), 774. https://doi.org/10.36922/ijocta026050021
Vancouver
Tarek Z, Hasan E. Enhanced audio classification with quantum-inspired neural layers and exponentially weighted attention fusion. An International Journal of Optimization and Control: Theories & Applications (IJOCTA). 2026;16(3):774. doi:10.36922/ijocta026050021
BibTeX
@article{Tarek2026,
author = {Tarek, Zahraa and Hasan, Esraa},
title = {Enhanced audio classification with quantum-inspired neural layers and exponentially weighted attention fusion},
journal = {An International Journal of Optimization and Control: Theories & Applications (IJOCTA)},
year = {2026},
volume = {16},
number = {3},
pages = {774},
doi = {10.36922/ijocta026050021}
}
EUVIMED