The code works in all Word versions. If the text is part of an image, it is no longer text, but pixels in an image; in that case, you need OCR software, not a macro.

Since the code you posted only concerned textboxes that were shape objects, that is what I showed you how to extend; you did not mention the need to process inlineshapes. The mere fact the code reports all the extracts as being from textboxes is inconsequential; outputting each & every shape type as well would require considerably more code - there are 24 different shape types.

WordArt is normally not a Shape object but an InlineShape object; if you want to extract such text, you need to process InlineShape objects as well. I suggest you have a go at that - it's not a major undertaking.

__________________
Cheers,
Paul Edstein
[Fmr MS MVP - Word]