The welding process's sound is a consequence of the modulation in amplitude of the current for the voltage of the electric arc. Previous experiments demonstrated that an experienced welder has an absolute dependence of the acoustics in the control of the welding process. In this work a new technique of detection of welding defects is presented based on the stability of the airborne acoustics of the process MIG/MAG for the transfer for short circuit mode. Statistical parameters of the acoustic pressure were determined and of the level of acoustic pressure for welding without defects from multiple tests; with these results have been development an algorithm of defects detection based on a mobile window that moves through the statistical signals calculated from the acoustics process. Finally a group of test plates within and without defects were tested; the algorithm based on the airborne acoustics of the developed process, showed satisfactory results for the detection of defects.