Electrical discharge machining (EDM) is one of the most accurate methods for machining conductive materials and has a number of important applications. In the EDM process the occurrence of electric charges between cathode and anode is accompanied by vibroacoustic signals, which can be used to develop highly efficient control and diagnostics systems. Experimental studies and modelling of the dynamic system of the EDM process carried out in this study show that parameters of acoustic signals can be used to estimate the current productivity and risks of the tool-electrode breakage and to optimize the tool feed rate. The obtained results of allows using acoustic signals in the control system of the tool electrode feed rate to prevent its breakage, and also setting the interelectrode gap to maximum productivity.