Make a scanned PDF searchable
Runs on your device — nothing is uploadedA scanned PDF looks like a document but behaves like a photo — you cannot search it, select it or copy from it. This reads each page with OCR and puts the recognised words back as a text layer, so the scan looks the same and behaves like real text.
How it works
- Add the scanned PDFDrop the scan. Each page is read as a picture, so a digital PDF with text already in it does not need this — you can copy that text directly.
- Make it searchablePress the button. Each page is rendered as an image and recognised in English, Sinhala and Tamil together. The progress shows which page is being read.
- Download the searchable copyYou get the same scan with a text layer you can select, search and copy. The pictures are untouched — only a layer of words goes under them.
What searchable means — and what it does not change
A searchable PDF is a picture of the page with words placed invisibly underneath it. The page you see is exactly the scan that went in — crest, letterhead, handwriting around the edges — because the picture is carried as it is. The text layer is only the convenience: searching finds the words, you can select and copy them, and a screen reader can read them.
Quality is not spent: the scan is not re-compressed and the resolution is kept. Two things do not happen: text is not rewritten into editable paragraphs, and columns are not rearranged — a page with two blocks of text side by side keeps the picture's order. If you need an editable rewrite, the recognised text is also shown in a box you can copy into Text to PDF.
When you do not need this at all — and when you do
Try selecting a word in your PDF reader before anything else. If a word highlights, the document already has text and there is nothing to add — its words can already be copied exactly. That covers every digital PDF (exported from a word processor, a browser or this site).
If the whole page highlights as a single block, or nothing highlights at all, the page is a picture and recognition is the right tool. A flatbed scan, a phone photo of a document, and a fax saved as PDF all behave this way. This tool turns that picture into a picture with words underneath it.
Accuracy — and why the photograph matters most
Recognition reads the shape of each letter, so a clean, straight, well-lit scan comes back close to perfect and a dim, crooked phone photo does not. On a clean 300 dpi English page our own tests produce no character errors; on the same page photographed at low resolution through a heavy JPEG, errors rise sharply. Same engine, same words — the picture decides.
That is also why Sinhala and Tamil work here: the same engine and the same fonts already read printed Sinhala and Tamil photographs elsewhere on this site, and a scanned PDF is only a set of such photographs in one file. Handwriting in any language is not readable — that is a different and much harder problem.
Questions
Is my scan uploaded?
No. Each page is rendered and recognised in your browser tab and never sent anywhere. You can verify in the network tab, or disconnect after the page loads and watch it still work once the models are cached.
My PDF already has text. Will this make it worse?
Nothing about the picture changes. A page that already had text keeps that text and gains a layer it did not need; a scanned page gains the ability to be searched. The tool reports which pages already had text so you can tell.
Does it work on handwritten scans?
No. Only printed or typed text — screenshots, scans, sharp phone photos of print. Handwriting in any language is not reliably readable.
How long does it take?
The first time, the engine and the language models are downloaded (about 6 MB for English, more if the pages turn out to be Sinhala or Tamil) — once, and cached afterwards. After that a few seconds per page is typical.
Can I correct the words before saving?
The recognised text for each page is shown in a box you can edit and copy. If you paste it into Text to PDF it becomes a fresh, editable document — but the searchable PDF itself keeps the picture as it was.
Does the file get a lot bigger?
Only a little. The scan's pictures stay as they are; the new text layer is just coordinates and characters, which is tiny compared to the images.
Related guides
Other tools
Merge PDF files
Combine several PDFs into one file, in the order you choose.
Split PDF files
Break one PDF into separate files, or pull out only the pages you need.
Delete, reorder and rotate PDF pages
Delete, reorder, rotate and duplicate pages in a visual grid.
Edit PDF files
Add text, a signature, white-out, a watermark or page numbers.
Compress PDF files
Cut file size for email and upload limits — three levels of compression.
Convert PDF to images
Export every page as a PNG or JPG, delivered as a ZIP.
Convert images to PDF
Turn JPG, PNG or WEBP images into a single PDF document.
Rotate PDF pages
Fix sideways or upside-down scans, for all pages or a few.
Extract text from an image
Read the words out of a photo or a screenshot, without retyping them.
Turn text into a PDF document
Type or paste a page and get it back as a laid-out document, not a text dump.
Extract text from a PDF
Pull the words out of a PDF as text you can copy — no upload, no account.
Add page numbers to a PDF
Number every page, in the corner you choose and the format you choose.
Add a watermark to a PDF
Stamp DRAFT, CONFIDENTIAL or your own words across every page.
Crop a PDF
Cut the white margins off a scan, or keep only the part of the page that matters.
Convert HEIC to PDF
Convert iPhone HEIC photos to PDF or JPG — no upload, no app.
Redact a PDF permanently
Permanently black out text — truly remove it, not just cover it.
Fill out a PDF form
Fill out PDF forms in your browser — text, checkboxes, signatures.