The invention relates to a method for the sound processing of a stereophonic signal inside a motor vehicle. In a first implementation (“driver” mode) the stereophonic sound source is centered in the middle of the dashboard for the ‘driver’ listen position. For this purpose, delays (t1-t4) are introduced into the frequency bands of the channels transmitted by the speakers, such that the driver appears to be at the center of a circle on which the car speakers are positioned. In a second implementation (“all passengers” mode), the phases of the signals of the two front channels are equalized, such that the sound source appears to be centered on the driver and the front passenger of the vehicle.