Salvaging text from a PDF back to InDesign

Question

As many of you know, Acrobat Pro has an Export feature that can pull the text out of a PDF. While trying to rebuild a lost publication, I exported the text. Acrobat does a nearly-great job but has one fault:
Textual words that break at a syllable end up in the docx text as two misspelled words.
For example, flowing in the text are words that are broken like:
muscu lar
overwhelm ing
It is just a spacebar space, yet both in Word and in InDesign, the spell-checker sees two misspelled words ... and doesn't offer to join them. Spell Checker sees them, but is brutally stupid about knowing what to do.
 
I have miles and miles of text, and this sort of thing is in every paragraph. What should I do to fix them speedily?

leo.r · Accepted Answer

Try to run it through chatGPT?

request in chatGPT:

combine broken words correctly: "muscu lar overwhelm ing"

result:

"muscular" "overwhelming"

You may need to make a more specific and detailed request to get better results for your entire text.

Mike Witherell · Answer

Thanks Leo. I just got back from trying the very same idea. ChatGPT fixed probably 95% of the mid-sentence linebreak errors, and also does well at rejoining syllable fragments. I had to tell it explicitly not to rewrite the text; just fix the breaks.

I also tried it with Google Gemini, and that AI engine could not comprehend what I was asking it to do. The limitation of ChatGPT appears to be about a 2,000 word limit. Still, the possibility of ChatGPT as a text-editing work-around is very encouraging!

Sign up

To post, reply, or follow discussions, please sign in with your Adobe ID.

Sign in to Adobe Community

To post, reply, or follow discussions, please sign in with your Adobe ID.

Scanning file for viruses.

This file cannot be downloaded