Automatic soundscape quality estimation using audio analysis

The huge growth of population size along with all the accompanying impacts, like traffic flow, commercial and industrial activities have led to a respective increase of noise pollution in the urban environments. In most cases, noise pollution in big cities is characterized by low-frequency and continuous background sounds. This ever-growing environmental problem engages health risks and major complaints of annoyance on behalf of millions of citizens. Therefore, sustainable urban planning needs to seriously take into consideration the task of mitigating environmental noise. In addition, the quality of the acoustic environment plays an important role in urban as well as in rural and natural spaces, since it has been proven to affect biodiversity. In this paper, we demonstrate how efficiently assessing soundscape quality can be applied to real recordings from various sites. The evaluation of the qualitative attributes of the soundscape is carried out combining space-sound-human presence. The mapping of the extracted feature statistics to the perceived soundscape quality level is achieved through a Support Vector Machine Regression model. Extensive experiments have been carried out on a real-world dataset and the resulting performance evaluation proves that the proposed architecture can be applied to assess the soundscape quality of both natural and urban spaces.