Statistical Processing of Subjective Test Data for Sound Quality Evaluation of Automotive Horn

Mahanta, T K and B, Venkatesham and Karri, Badarinath (2015) Statistical Processing of Subjective Test Data for Sound Quality Evaluation of Automotive Horn. In: National Symposium on Acoustics “Acoustics for Ocean Environment” (NSA), October 7-9, 2015, Goa, India.

Preview

Text
NSA_2015.pdf - Published Version
Download (466kB) | Preview

Abstract

Product sound quality has a significant role in buying decision and customer satisfaction. An often used method to assess the sound quality of any product or equipment is a subjective listening test where the sound is heard by a panel of subjects (jury) who then rate the sound quality. Subjects use a semantic differential rating wherein they evaluate the presented sounds based on a bipolar variable. The two extremes of the rating scale are labeled with an adjective and its antonym respectively. In the present study, a subjective listening test has been conducted to assess sound quality of automotive horns. The data obtained are then analyzed using statistics to gain insights. Twenty two horn sound samples were judged by thirty participants aged 20-40 years who had normal hearing. Binaural head set (BHS) instrument was used to record horn sound samples in open ground (neglecting wind noise effect). Sounds are recorded two meter from horn in front direction and used for subjective test. For the subjective test and subsequent statistical analysis, a four step procedure has been used. In the first step, the participants were asked to rate the sound quality for each horn based on seven bipolar variables. These bipolar variables are soft/loud, calm/frightening, slow/fast, relax/tense, safe/danger, vague/distinct and pleasant/unpleasant. For each bipolar variable, a seven verbal interval scale was used ranging from one extreme to another in degree, for example extremely pleasant to extremely unpleasant.

[error in script]

IITH Creators: