ColabSomGeo: Ferramenta de monitoramento de poluição sonora subsidiada por modelo de aprendizagem

• Sannyer Cardoso Carvalho Nery IFAM
• Fani Tamires de Souza Batista IFAM
• Sergio Augusto Coelho Bezerra IFAM

DOI: https://doi.org/10.5753/webmedia_estendido.2025.16410

Resumo

Sound pollution in urban centers is a public health issue, with excessive noise negatively impacting the physical and mental health of the population. Traditional monitoring methods are often costly, making widespread data collection difficult. This paper introduces ColabSomGeo, a low-cost, scalable web tool for visualizing and analyzing georeferenced sound pollution data. Data collection is performed collaboratively through a Telegram chatbot, where users submit audio recordings and their locations. A server application processes the data, classifies the noise type using a machine learning model, and stores the results. The tool aims to assist public managers in formulating policies, support scientific research, and raise public awareness about the impacts of noise on well-being. The code is available at: https://github.com/colabsomgeo

Palavras-chave: Poluição sonora, classificação de sons urbanos, sistemas colaborativos, georreferenciamento, monitoramento, Restful, API, Chatbot

Referências

C. A. R. Beltran et al. 2019. Plataforma de aprendizado de máquina para detecção e monitoramento de alunos com risco de evasão. (2019), 1591.

Brüel & Kjær. 1996. Environmental Noise. Technical Review, No. 3.

Ana Mercês Nicolaci da Costa and Mariano Pimentel. 2011. Sistemas colaborativos para uma nova sociedade e um novo ser humano. In Sistemas Colaborativos (1 ed.), Mariano Pimentel and Hugo Fuks (Eds.). Vol. 1. Elsevier, Rio de Janeiro, 3–15.

Samir N. Y. Gerges. 2000. Ruído: Fundamentos e Controle (2 ed.). NR Editora, Florianópolis.

Numpy Developers. [n. d.]. NumPy. [link]. Acesso em: 19 de agosto de 2023.

Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas Kopf, Edward Yang, Zachary DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu Fang, Junjie Bai, and Soumith Chintala. 2019. PyTorch: An Imperative Style, High-Performance Deep Learning Library. In Advances in Neural Information Processing Systems 32, H. Wallach, H. Larochelle, A. Beygelzimer, F. d’Alché Buc, E. Fox, and R. Garnett (Eds.). Curran Associates, Inc., 8024–8035.

Mirco Ravanelli, Titouan Parcollet, Peter Plantinga, Aku Rouhe, Samuele Cornell, Loren Lugosch, Cem Subakan, ASR Gradient, Hwidong Na, Vimal Panayotov, et al. 2021. SpeechBrain: A general-purpose speech toolkit. In arXiv preprint arXiv:2106.04624.

Thomas D. Rossing, F. Richard Moore, and Paul A. Wheeler. 2014. The Science of Sound (3 ed.). Pearson, San Francisco.

A. S. Santos et al. 2020. Desafios e oportunidades da aplicação de Sistemas Ciberfísicos no monitoramento da poluição urbana. (2020).

Igor Sabarense Coelho Silva and Felipe Cunha. 2022. Monitoramento da Poluição Sonora em Belo Horizonte utilizando Redes de Sensoriamento Participativo. Trabalho acadêmico.

World Health Organization. 2018. Environmental Noise Guidelines for the European Region. Technical Report. WHO Regional Office for Europe, Copenhagen. [link]

Binyu Zhao and Jian Liang. 2024. Hierarchical-Concatenate Fusion TDNN for sound event classification. Plos one 19, 10 (2024), e0312998.