Urban background sounds under various weather conditions categorized for virtual acoustics.

One of the major problems in the auralization of complex urban scenarios is the creation of ambience, i.e. the diffuse background which represents the sound floor on which the acoustic scene of interest is created. Omitting a realistic ambience results in a sterile and implausible impression. To overcome this drawback, a hybrid approach with admixture of spatially recorded background sounds seems to be a promising approach. This paper provides a protocol to record and categorize background soundscapes for such applications. The freely accessible database contains 102 calibrated recordings at the IHTApark in Aachen (Germany) across various weather conditions, segmented into calibrated 30-second snippets. The categorization is performed based on the meteorological and acoustic parameters.