The present invention relates to a method of adjusting the sound levels originating from several audio sources, within a television receiver. The audio components originating from at least two audiovisual or audio sources are received in an audiovisual receiver and mixed so as to generate a single audio signal. The levels assigned to each source can be adjusted by the user with the aid of an icon displayed on a screen. The icon comprises at least two points representing the two sources and a cursor which moves along two dimensions, the distance between this cursor and the point representing a source being proportional to the sound level assigned to this source. The present invention also relates to the audiovisual terminal for the implementation of the method.