The invention discloses a vehicle-mounted
sound system fine calibration method based on a
critical frequency band and a
stereo imaging sound layer model. The method is based on an innovative'imaging sound layer 'psychological
acoustic model, and the core of the method comprises the following steps: collecting a dual-track
acoustic response in a vehicle; based on an ERB model, dividing an analysis sub-
frequency band corresponding to an imaging sound layer, and identifying a miscoordination target
frequency band by calculating an energy difference; and asymmetrically adjusting the parameters of the other sound channel by taking a single sound channel as a fixed reference so as to reduce the deviation. According to the method, an automatic progressive double-ring calibration strategy is preferably adopted, correction is completed in combination with a
sine wave probe and a fine step length, and the
sound field quality is ensured through objective
verification and subjective acceptance. In addition, a parameter prediction model optimization
mechanism based on an expert sample
library is introduced, so that the
system has a
continuous optimization capability. According to the method, the calibration process is deeply aligned with the auditory characteristics of human ears, the problem of
sound image blurring in the complex acoustic environment of the automobile cabin is effectively solved, and the immersive high-fidelity experience is remarkably improved.