MSA-TCN: Robust Urban Suspicious Sound Detection Using Multi-Scale Temporal Convolutions and Dual Attention
DOI:
https://doi.org/10.62411/faith.3048-3719-313Keywords:
Acoustic signal processing, Audio event detection, Deep learning for audio, Edge AI for sound analysis, Multi-scale temporal convolutional networks, Sustainable Cities and Communities, Suspicious sound detection, Urban sound classificationAbstract
Efficient suspicious sound event detection is essential for urban surveillance, smart-city monitoring, and public safety systems; however, many existing deep learning approaches suffer from high computational complexity, unstable optimization, and limited suitability for real-time deployment. This paper proposes a Multi-Scale Attention-based Temporal Convolutional Network (MSA-TCN) for robust urban sound classification. The model operates on MFCC representations and integrates dilated multi-scale temporal convolutions to capture both short impulsive events and long-duration acoustic patterns. To enhance training stability and generalization, spectral normalization is applied to convolutional layers, while channel and spatial attention mechanisms are incorporated to adaptively emphasize discriminative temporal–spectral features. The proposed MSA-TCN is evaluated on the UrbanSound8K dataset using stratified 80/10/10 train–validation–test splits and compared against Dense and LSTM baseline models under identical experimental conditions. MSA-TCN achieves a test accuracy of 92.34% and a macro-averaged F1-score of 0.92, outperforming the Dense and LSTM baselines by 3.48% and 2.09%, respectively, while using only 128K parameters. Comprehensive ablation studies demonstrate that spectral normalization and multi-scale temporal modeling are the most critical components, each contributing approximately 2–2.5 percentage points to test accuracy and significantly improving convergence stability. Attention mechanisms primarily enhance training efficiency and class-specific discrimination rather than absolute accuracy. Confusion-matrix and per-class analyses confirm strong recall for safety-critical events such as gunshots and sirens, while training dynamics reveal stable convergence and efficient learning. These results validate MSA-TCN as a lightweight, robust, and effective framework for real-time urban sound classification in public safety applications.
Downloads
References
Z. Momynkulov et al., “Fast Detection and Classification of Dangerous Urban Sounds Using Deep Learning,” Comput. Mater. Contin., vol. 75, no. 1, pp. 2191–2208, 2023, doi: 10.32604/cmc.2023.036205.
C. Castorena, M. Cobos, J. Lopez-Ballester, and F. J. Ferri, “A safety-oriented framework for sound event detection in driving scenarios,” Appl. Acoust., vol. 215, p. 109719, Jan. 2024, doi: 10.1016/j.apacoust.2023.109719.
A. F. R. Nogueira, H. S. Oliveira, J. J. M. Machado, and J. M. R. S. Tavares, “Sound Classification and Processing of Urban Environments: A Systematic Literature Review,” Sensors, vol. 22, no. 22, p. 8608, Nov. 2022, doi: 10.3390/s22228608.
A. Shailendra, C. Bengani, K. S. Kumari, P. Senthilraja, A. Prithivi, and S. Ramesh, “Suspicious Activity Detection based on Audio Detecting Methodology using Deep Learning,” in Recent Trends in Data Science and its Applications, 2023, pp. 683–687. doi: 10.13052/rp-9788770040723.131.
F. Lo Scudo, E. Ritacco, L. Caroprese, and G. Manco, “Audio-based anomaly detection on edge devices via self-supervision and spectral analysis,” J. Intell. Inf. Syst., vol. 61, no. 3, pp. 765–793, Dec. 2023, doi: 10.1007/s10844-023-00792-2.
M. Neri, F. Battisti, A. Neri, and M. Carli, “Sound Event Detection for Human Safety and Security in Noisy Environments,” IEEE Access, vol. 10, pp. 134230–134240, 2022, doi: 10.1109/ACCESS.2022.3231681.
E. O. Ibam and A. E. Mesioye, “Decoupling Cyber Sabotage in Industrial Telemetry from Mechanical Faults Using Gradient-Based Forensic Edge XAI,” J. Futur. Artif. Intell. Technol., vol. 3, no. 1, pp. 1–18, Mar. 2026, doi: 10.62411/faith.3048-3719-322.
S. Chandrakala, A. Pidikiti, and P. V. N. Sai Mahathi, “Spectro Temporal Fusion with CLSTM-Autoencoder based approach for Anomalous Sound Detection,” Neural Process. Lett., vol. 56, no. 1, p. 39, Feb. 2024, doi: 10.1007/s11063-024-11485-4.
H. Zhang, M. Wu, X. Cai, and W. Liu, “Scene-dependent sound event detection based on multitask learning with deformable large kernel attention convolution,” PLoS One, vol. 20, no. 5, p. e0322002, May 2025, doi: 10.1371/journal.pone.0322002.
P.-A. Grumiaux, S. Kitić, L. Girin, and A. Guérin, “A survey of sound source localization with deep learning methods,” J. Acoust. Soc. Am., vol. 152, no. 1, pp. 107–151, Jul. 2022, doi: 10.1121/10.0011809.
L. Marchegiani and P. Newman, “Listening for Sirens: Locating and Classifying Acoustic Alarms in City Scenes,” IEEE Trans. Intell. Transp. Syst., vol. 23, no. 10, pp. 17087–17096, Oct. 2022, doi: 10.1109/TITS.2022.3158076.
M. Y. Shams, T. Abd El-Hafeez, and E. Hassan, “Acoustic data detection in large-scale emergency vehicle sirens and road noise dataset,” Expert Syst. Appl., vol. 249, p. 123608, Sep. 2024, doi: 10.1016/j.eswa.2024.123608.
J. Salamon, C. Jacoby, and J. P. Bello, “A Dataset and Taxonomy for Urban Sound Research,” in Proceedings of the 22nd ACM international conference on Multimedia, Nov. 2014, pp. 1041–1044. doi: 10.1145/2647868.2655045.
K. Zaman, M. Sah, C. Direkoglu, and M. Unoki, “A Survey of Audio Classification Using Deep Learning,” IEEE Access, vol. 11, pp. 106620–106649, 2023, doi: 10.1109/ACCESS.2023.3318015.
L. Banchero and J. J. López, “Designing a practical physical framework for AI-enhanced sound detection and localization in vehicles,” INTER-NOISE NOISE-CON Congr. Conf. Proc., vol. 270, no. 8, pp. 3333–3344, Oct. 2024, doi: 10.3397/IN_2024_3311.
V.-T. Tran and W.-H. Tsai, “Acoustic-Based Emergency Vehicle Detection Using Convolutional Neural Networks,” IEEE Access, vol. 8, pp. 75702–75713, 2020, doi: 10.1109/ACCESS.2020.2988986.
Y. Gong, Y.-A. Chung, and J. Glass, “AST: Audio Spectrogram Transformer,” in Interspeech 2021, Aug. 2021, pp. 571–575. doi: 10.21437/Interspeech.2021-698.
M. Zohaib, M. Asim, and M. ELAffendi, “Enhancing Emergency Vehicle Detection: A Deep Learning Approach with Multimodal Fusion,” Mathematics, vol. 12, no. 10, p. 1514, May 2024, doi: 10.3390/math12101514.
A. Vaswani et al., “Attention Is All You Need,” arXiv, vol. 30, Aug. 2023, [Online]. Available: http://arxiv.org/abs/1706.03762
V. Kumar and A. Purohit, “Tuned Attention-Guided Fusion of Deep and Handcrafted Features for Distinguishing AI-Generated and Human-Created Artworks,” J. Futur. Artif. Intell. Technol., vol. 2, no. 4, pp. 629–647, Jan. 2026, doi: 10.62411/faith.3048-3719-291.
Jingdong Chen, J. Benesty, Yiteng Huang, and S. Doclo, “New insights into the noise reduction Wiener filter,” IEEE Trans. Audio, Speech Lang. Process., vol. 14, no. 4, pp. 1218–1234, Jul. 2006, doi: 10.1109/TSA.2005.860851.
R. Lee, M.-S. Kang, B.-H. Kim, K.-H. Park, S. Q. Lee, and H.-M. Park, “Sound Source Localization Based on GCC-PHAT With Diffuseness Mask in Noisy and Reverberant Environments,” IEEE Access, vol. 8, pp. 7373–7382, 2020, doi: 10.1109/ACCESS.2019.2963768.
R. Munirathinam and S. Vitek, “Sound Source Localization and Classification for Emergency Vehicle Siren Detection Using Resource Constrained Systems,” in 2024 34th International Conference Radioelektronika (RADIOELEKTRONIKA), Apr. 2024, pp. 1–5. doi: 10.1109/RADIOELEKTRONIKA61599.2024.10524053.
K. Nugroho, I. H. Al Amin, N. A. Noviasari, and D. R. I. M. Setiadi, “Prosodic Spatio-Temporal Feature Fusion with Attention Mechanisms for Speech Emotion Recognition,” Computers, vol. 14, no. 9, pp. 1–15, 2025, doi: 10.3390/computers14090361.
J. F. Gemmeke et al., “Audio Set: An ontology and human-labeled dataset for audio events,” in 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Mar. 2017, pp. 776–780. doi: 10.1109/ICASSP.2017.7952261.
M. Cantarini, L. Gabrielli, A. Mancini, S. Squartini, and R. Longo, “A3CarScene: An audio-visual dataset for driving scene understanding,” Data Br., vol. 48, p. 109146, Jun. 2023, doi: 10.1016/j.dib.2023.109146.
N. N. Wijaya, D. R. I. M. Setiadi, and A. R. Muslikh, “Music-Genre Classification using Bidirectional Long Short-Term Memory and Mel-Frequency Cepstral Coefficients,” J. Comput. Theor. Appl., vol. 1, no. 3, pp. 243–256, Jan. 2024, doi: 10.62411/jcta.9655.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Micah Shallom Bawa, Shehu Mohammed Yusuf, Sani Saleh Saminu
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
