A Multiple Regression Model for Predicting Loudspeaker Preference Using Objective Measurements: Part I - Listening Test Results
Part I of this paper presents the objective measurements and listening test results on 13 loudspeakers rated according to preference, spectral balance and distortion. In part II the data provides the framework for the development and verification of a multiple regression model that predicts listeners? preferences based on objective measurements. We review relevant predictive models and test one model currently used by Consumers Union (CU), a consumer product testing organization in the United States. There is no correlation between listeners? loudspeaker preference ratings and CU?s predicted accuracy scores (r = 0.05; p = .81). As the CU model is based largely on the loudspeaker?s 1/3-octave sound power response we conclude that measured sound power, alone, cannot accurately predict its perceived sound quality.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is temporarily free for AES members.