In response to the demand for non-invasive measurement of the dynamic characteristics of AC contactors, this study proposes a sound signal-based method for measuring the dynamic parameters of AC contactors by analyzing and researching the sound mechanism of the contactor action process and taking into account the characteristics of sound signals. The method uses ensemble empirical mode decomposition to adaptively decompose and select sensitive components of the sound signal. Furthermore, based on the kurtosis feature of the signal-sensitive components, the acoustic event boundaries in the complete sound signal are accurately segmented and calibrated, thereby determining the key event nodes of the contactor operation and achieving indirect measurement of the dynamic characteristic parameters of the contactor. Experimental results show that this method can effectively measure the dynamic characteristic parameters of the contactor, laying a foundation for further research on the operating status of the contactor.