Systems and methods of audio reproduction are described. In one embodiment, one or more processors can be configured, for example, to receive an L channel signal and an R channel signal, to combine the L channel signal and the R channel signal, to average the combined signal, to detect an impulse signal in the averaged signal, to multiply the impulse signal, to scale the multiplied impulse signal, and to add the scaled impulse signal to the averaged signal to form a resultant signal. Sound can be generated, for example, by speakers based on at least the resultant signal.