Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{193738,
        author = {Mrs .H Teja and P Sree Nandini and V Poojitha and Y Surendra},
        title = {Residual CNN-Based Audio Classification of Emergency Sirens for Smart Traffic Control},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {2224-2231},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=193738},
        abstract = {Emergency vehicle sound classification is an important component of intelligent traffic management systems, as it helps reduce delays for emergency services during critical situations. This project proposes a method for detecting and classifying emergency vehicle siren sounds using Residual Convolutional Neural Networks (CNNs). The system utilizes a dataset of WAV audio files containing siren sounds from ambulances, fire trucks, and police vehicles. Mel-Frequency Cepstral Coefficients (MFCCs) are extracted as key audio features to capture the unique frequency patterns of emergency sirens. Audio signal processing techniques, including noise reduction and feature normalisation, are applied during preprocessing to improve model performance. The extracted MFCC features are then used to train a residual CNN model that accurately classifies the siren sounds into their respective categories. Experimental results show that the proposed model achieves high classification accuracy, demonstrating its reliability for real-time applications. By integrating the classification system with dynamic traffic signal control, traffic lights can be automatically adjusted to provide priority access to approaching emergency vehicles. This approach enhances overall traffic efficiency and reduces emergency response time. The proposed system shows strong potential for deployment in smart city environments. In the future, the system can be combined with live traffic data to improve traffic management in cities.},
        keywords = {Emergency vehicle sound classification, residual convolutional neural networks (CNN), mel-frequency cepstral coefficients (MFCC), intelligent traffic management, dynamic traffic signal control, audio signal processing.},
        month = {March},
        }