A Method for Automatic Whoosh Sound Description

Usually, a sound designer achieves artistic goals by editing and processing thepre-recorded sound samples. To assist navigation in the vast amount of sounds,the sound metadata is used: it provides small free-form textual descriptions ofthe sound file content. One can search through the keywords or phrases in themetadata to find a group of sounds that can be suitable for a task.Unfortunately, the relativity of the sound design terms complicate the search,making the search process tedious, prone to errors and by no means supportive ofthe creative flow. Another way to approach the sound search problem is to usesound analysis. In this paper we present a simple method for analyzing thetemporal evolution of the ``whoosh'' sound, based on the per-band piecewiselinear function approximation of the sound envelope signal. The method usesspectral centroid and fuzzy membership functions to estimate a degree to whichthe sound energy moves upwards or downwards in the frequency domain along theaudio file. We evaluated the method on a generated dataset, consisting of whitenoise recordings processed with different variations of modulated bandpassfilters. The method was able to correctly identify the centroid movementdirections in 77% sounds from a synthetic dataset.