In previous papers, the authors presented a new technique of identifying a nonlinear multi-degree-of-freedom vibratory system and showed its applicability by numerical simulation. To see how an actual vibratory system is identified by the presented technique, an identification system is developed based on the technique and using a personal computer, and is applied to some simple vibratory systems. The identification results is then evaluated by comparing the actual response of the vibratory systems with the predicted response using the identified parameters. It is shown in this way that the developed technique can well be used for identifying the vibratory systems in a reasonable time.