A scanned PDF is a stack of pictures — nothing to search, nothing to copy. This lays an invisible text layer under the images. It looks exactly the same, but now you can search it.
Try searching for a word in your reader when it finishes.
This one needs an upload, and is deleted afterwards
Browsers cannot do this calculation, so the file goes to our server. It is deleted automatically 60 minutes after the job finishes, and the working copy lives in a memory filesystem that is wiped whenever the server restarts.
We do not log file names and do not log file contents — that is a hard rule in our engineering standards, not a setting.
What it does — and what it doesn't
The appearance is unchanged. The text is invisible, underneath the picture, and never covers what you can see.
Accuracy depends on the original. Clean, high-contrast scans do very well; dim, skewed or small print produces errors. Run Remove scan grey first — it makes a large difference.
Handwriting is not recognised. Pages full of table rules often turn lines into characters.
OCR is one of the slower jobs; long documents take a while and show progress.
Common questions
How is this different from OCR text recognition?
That one hands you the text as a file. This one puts the text back into the PDF, so the file is still your document — just searchable.
Do recognition errors show up?
No — the layer is invisible, so mistakes affect searching, not appearance. If a word cannot be found, it was probably misread.
Is it safe to use as an official document?
The visible content is unchanged, so the original's standing is unaffected. But do not treat the OCR text as the authoritative content — it is a search aid.
