An image processing device includes a displayer; and a hardware processor that obtains voice recognition data that is a voice recognition result related to a voice vocalized in a state in which at least one operation screen is displayed in the displayer, determines a search target character string on the basis of the voice recognition data, executes search processing of searching for one voice operation command that agrees with the search target character string among a plurality of voice operation commands including a voice operation command group related to a first screen related to the image processing device, and a voice operation command group related to a second screen displayed according to user's operation for the first screen, and executes processing corresponding to the one voice operation command that has been searched for by the hardware processor.