Make a scanned PDF searchable
Add a hidden text layer to scanned pages so you can search and copy the text. Pages that get a text layer keep their original content; pages you choose to redo are rebuilt from an image of the page.
Your file is processed in this browser tab and is not uploaded. Up to 100 MB and 1000 pages per file. Up to 100 pages processed per run.
This tool needs JavaScript. If it does not start, enable JavaScript in your browser and reload this page.
This browser cannot run the tool. Please use a current version of Chrome, Edge, Firefox or Safari.
This tool runs on desktop browsers only for now. Text recognition needs more memory than phones and tablets reliably have, and it has not been tested on them yet. Please use a computer.
No file selected yet.
Choose the document's languages, not the language of this page.
Some pages have only a little text (fewer than 50 characters) — often just a page number or a broken OCR layer. Decide for each: add a text layer, redo the page, or skip it.
| Page | Existing text | Plan |
|---|
Result
| Page | Existing text | Result |
|---|
How to use it
- Choose a scanned PDF and tick the document languages.
- Check the plan for each page: pages that already have enough text are skipped; pages with little text are yours to decide.
- Select Make searchable and save the new PDF.
What it does
- Recognises the text of each chosen page and lays it over the page as invisible text, placed on the words — the visible page is not changed.
- Skips pages that already have text by default, and shows how much text each page had.
- "Redo page" replaces a wrong or partial old text layer: that page becomes an image of itself with a new text layer. "Redo all pages" marks every page for this. One run processes at most 100 pages: for a longer file, split it into parts (Split PDF), make each part searchable and merge them again (Merge PDF).
- Lists processed, skipped and failed pages; a page that fails stays as it was.
What it does not do
- Thai documents are not offered: Thai tone marks are often lost in recognition, so searches would miss words. Use Scanned PDF to text and proofread the Thai text.
- It does not correct recognition errors; a misread word is searchable only as it was read. Check important pages, for example by copying their text.
- It does not run on phones and tablets yet. Use a computer with a current browser.
What the new file does not keep
- Redone pages: their old text layer, and some sharpness. The page is saved as a JPEG image at up to 300 dpi, lower for large pages (at most about 12 million pixels per page), so fine detail can blur and compression artefacts can appear.
- Digital signatures: the file is rewritten.
File limits
- Up to 100 MB and 1000 pages per file.
- Up to 100 pages processed per run.
Questions
- Why skip pages that already have text?
- They are usually searchable already. But having text does not prove the page is searchable — it may hold only a page number or a wrong old OCR layer — so the tool shows each page's text amount and lets you redo it.
- Does the file get bigger?
- A little for added text layers. Redone pages are stored as images and can make the file noticeably larger.
Privacy
- The file is read and processed by code running in this browser tab. It is not uploaded to our server or to any other service.
- The tool loads its own program files from this site; it does not contact other services while it works.
- Nothing is kept after you close the tab. Results exist only until you save them.