Application of machine learning in modeling the quality of acoustic environments: A review

In recent years, there has been a significant increase in the use of machine learning (ML) methods for modeling the acoustic environment quality. This review evaluates supervised, ensemble, and unsupervised ML models used to assess and predict acoustic environment quality. Artificial neural networks (ANNs) have been the most widely used ML model in this domain, while recent advancements have increased the adoption of techniques such as ensemble and deep learning. India led global publications in this field, with “equivalent continuous sound levels (L_eq)” and “A-weighted equivalent continuous sound levels (L_Aeq)” being the most extensively studied output parameters. Future research should focus on integrating advanced techniques to enhance predictive accuracy and implementing ML models to improve the management of acoustic environment quality.