OCR PDF
Recognize text in a scanned PDF and download a searchable PDF or plain text. Runs locally in your browser — free, instant, nothing uploaded. Accuracy depends on scan quality.
Related tools
Everything you need to OCR a scanned PDF
Real OCR, not just text extraction
Reads the actual pixels of a scanned page and recognizes the words in them — works on documents with no underlying text layer at all.
Searchable PDF output
Get back a PDF that looks exactly like your scan, but with an invisible text layer you can select, search, and copy from.
100% private — nothing uploaded
Recognition runs entirely in your browser via WebAssembly. Your file never touches a server, which matters for contracts, transcripts, or anything sensitive.
Honest about accuracy
OCR accuracy depends on scan quality — this page says so plainly instead of promising perfection it can't deliver on a blurry photo of a page.
Real per-page progress, with cancel
A multi-page scan can take a few minutes. The progress bar tracks actual pages completed, not a frozen placeholder, and you can cancel at any time.
Pick exactly which pages to OCR
Select specific pages from a thumbnail grid, or set a page range on longer documents, instead of always processing the whole file.
Three steps to a searchable PDF
- 1
Upload your scanned PDF
Drag and drop a scanned PDF onto the page or click to browse — old contracts, scanned forms, receipts, photographed pages, anything with no real text layer.
- 2
Pick which pages to OCR
By default every page runs. Select just the pages you need from the thumbnail grid, or set a page range on longer documents.
- 3
Download a searchable PDF or plain text
Click Run OCR. Recognition runs locally in your browser, page by page, with a live progress bar. Download a searchable PDF (the scan, with an invisible text layer added) or the recognized text as a .txt file.
Built for anyone stuck with unsearchable scans
Whether it's an old contract, a scanned form, or a photo of a printed page — PixelTools reads the text and gives you back a searchable PDF or plain text, free, with no upload.
Make old scanned contracts searchable
Turn a filing cabinet's worth of scanned agreements and forms into PDFs you can actually search and copy text from.
Digitize photographed pages
Recognize text in a phone photo of a book page, whiteboard, or printed document so you can copy it out instead of retyping it.
Archive paper records properly
Recognize text from scanned receipts, forms, and records before archiving so they're indexed and searchable, not just images.
Recover text from scanned research
Pull readable text out of scanned academic papers, articles, or handouts that predate digital text layers.
Frequently asked questions
Is this OCR PDF tool free?
+
Yes — completely free, no account required, no watermarks, no limit on how many PDFs you OCR.
Does my PDF get uploaded to a server?
+
No. Recognition runs entirely in your browser using tesseract.js, an open-source OCR engine compiled to WebAssembly. Your file is never sent anywhere — it stays on your device the whole time, which matters for contracts, medical records, or anything else you'd rather not upload.
How accurate is the text it recognizes?
+
It depends heavily on scan quality. A clean, straight, high-resolution scan (roughly 300 DPI) typically comes out very accurate. A blurry photo of a page, a skewed scan, or messy handwriting will produce more mistakes — OCR reads shapes, not meaning, so always double-check anything important before relying on it.
What's the difference between this and Extract Text from PDF?
+
Extract Text from PDF (/pdf-to-text) pulls text that already exists inside a PDF — it's instant, but only works if the PDF already has a real text layer. OCR PDF is for the opposite case: a scanned or photographed PDF that's really just pictures of text, with nothing to extract yet. This tool reads those pictures and recognizes the words in them, which takes longer but works where the other tool can't.
What is a "searchable PDF"?
+
It's your original scan with an invisible layer of recognized text placed over it. It looks identical to the scan, but you can now select, search, and copy text from it — and it's indexed properly if you search your files later.
How long does OCR take?
+
Longer than most tools on this site — recognition happens page by page, and a multi-page document can take a few minutes. The progress bar tracks real per-page progress, and you can cancel at any point if you change your mind.
Does it work on password-protected PDFs?
+
Not currently. Remove the password in your PDF reader first, then upload the unlocked file here.
What languages does it support?
+
English only for now. The recognition model is downloaded on first use — after that, re-running OCR on another PDF in the same session is faster.
OCR your scanned PDF now
Free, instant, no account needed.
OCR PDF →