1993-05-01

Quantification of Subjective Unpleasantness Using Roughness Level 931332

The subjective estimation of real car interior noise by a semantic differential method was conducted for determining major factors which affect sound qualities of car interior noise. As a result of principal factor analysis, three factors, i.e., pleasantness, powerful-ness and boomingness were extracted from the 14 pairs of adjectives.
When we consider for improving sound qualities of car interior noise, the most important factor among these three factors was pleasantness. we should try, therefore, to reduce harshness and unpleasantness of car interior noise by examining its time envelope patterns, frequency compornents and sound pressure level, which allowed us to determine what is the most effective method for obtaining good sound quality.
According to the precise investigation of the time envelope patterns of real car interior noise and artificial sounds, it was clarified that the degree of how the time envelope patterns of car interior noise was rough or smooth related well with the subjective feeling of being unpleasant and pleasant.
Finally we had the following conclusion, i.e., in order to quantify subjective unpleasantness, we had better to extract time varying nature of the envelope of the sound. To obtain this measure, we constructed a root mean square circuit with its time constant 20ms. The time signal of the sound pressure was fed to this circuit and the power spectrum of its output signal between 5Hz and 160Hz were divided by that of the refference sound and we take its logarithm and multiplied 10 to obtain “roughness level” in dB.
When we introduced this measure, we could find that the subjective estimation of unpleasantness of the sound was related well with this measure.

SAE MOBILUS

Subscribers can view annotate, and download all of SAE's content. Learn More »

Access SAE MOBILUS »

Members save up to 16% off list price.
Login to see discount.
Special Offer: Download multiple Technical Papers each year? TechSelect is a cost-effective subscription option to select and download 12-100 full-text Technical Papers per year. Find more information here.
X