This invention is an apparatus and a method for language training and general entertainment and role-playing. It employs a digital processing system that can play back audio or audio-visual program material, using special software to selectively direct the audio of this playback through the digital processing system's speaker system or through a headset, with the user taking the audio routed to her headset--in particular speech--as a model and prompt to repeat this speech audibly. The user plays the role of the person in the program material whose speech is being routed to her headset, and the special software further permits the user to record her dubbing efforts along with the program material, producing a version of the program material with the user's voice recorded in place of, or in conjunction with, the voice of a character in the program material.