My application is currently able to extract images from MS Word documents in Python by treating them as zipfiles and grabbing the images from /word/media/, like this:
