As smart voices gradually enter the human living environment, people can interact with controlled systems through simple commands without the need for a screen. Even people without specialist training can use voice sensing control to easily complete the operation of a robot arm while avoiding incorrect operation caused by insufficient professional knowledge. In this study, we use the Google speech recognition engine to convert input speech into text information, and then perform speech recognition, word segmentation, and part-of-speech (POS) tagging. We also propose an improved rapid automatic keyword extraction (IRAKE) method, which uses the word string matching feature in the dictionary method to correspond to the relevant execution action function. The experimental results show that the accuracy rate of the proposed IRAKE method is 16% higher than that of the traditional rapid automatic keyword extraction (RAKE) method.