Perceptual Dimensions of Fan Noise and Their Relationship to Indexes Based On the Specific Loudness
Stephan Töpken
Steven Van De Par
Abstract
Fan noise is a typical part of environmental noise perceived by humans in their daily life. In order to successfully develop more pleasant fan sounds, it is desirable to understand and characterize perceptually relevant aspects of fan noise beyond current technical measures like the
dB(A). The aim of this study is a determination of the perceptual dimensions of fan noise, an identification of those dimensions which are relevant for the (un-)pleasantness and the development of (psycho-)acoustic indexes reflecting the most relevant perceptual dimensions. In a first experiment,
a semantic differential was developed based on a free verbalization of adjectives and a rating of their suitability. In a second experiment, 35 different fan noises were rated by 45 participants using the developed semantic differential consisting of 29 adjective scales. As a result of a principal
component analysis (PCA), six perceptual dimensions could be identified as "pleasant", "humming", "shrill", "monotone", "reverberant" and "noise-like". Another PCA revealed five groups of sounds which differed systematically in terms of their semantic profiles and calculated specific loudness.
Based on an analysis of the specific loudness patterns of the signals, two new indexes were derived that differ systematically among three major groups of ventilator sounds comprising of unpleasant, humming and pleasant sounds.