Leisure and road traffic are two of the major contributors to environmental noise in urban agglomerations. This environmental problem is particularly important in some cities, where there is a massive circulation of traffic vehicles and numerous outdoor social activities with a leisure noise mainly composed by human-related sounds. Due to the apparent differences between human-related sounds and road traffic sounds, the descriptors used to approach a proper evaluation of leisure noise may not be the same as the ones used for road traffic noise evaluation. This paper is aimed at analyzing the road traffic and leisure noise characteristics and the most important parameters to study and evaluate both noise sources. Therefore, as usually both noise sources are mixed, a model has been developed and tested which, when provided an environmental noise measurement, will be able to identify what kind of noise (leisure or road traffic noise) is dominant in a given environment. This model, done by using Support Vector Machines (SVMs) and Regression Analysis algorithms, allows focusing the abatement actions towards the noise source identified as dominant, and thus, addressing a better management and a proper control of the environmental noise in urban environments.