Load a scanned PDF and read its text with on-device OCR. You get a searchable PDF and the extracted text — nothing is uploaded.
Drop a PDF here, or click to browse
.pdf · stays on your device
How to OCR a PDF
Three steps, all in your browser.
1
Add your scanned PDF
Drag in a PDF or click to browse. Image-based (scanned) PDFs are exactly what OCR is for.
2
Run OCR
Click Run OCR. The English text-recognition engine loads once, then reads each page — progress shows as it goes.
3
Download
Get a searchable PDF (looks the same, but the text is selectable) and the extracted plain text as .txt.
Once the text layer exists you can pull it out with PDF to Text, or move it into a document with PDF to Word. To render the pages as pictures instead, use PDF to JPG.
Frequently Asked Questions
How do I OCR a PDF?
Load a scanned PDF and click Run OCR. The tool renders each page, recognises the text with an on-device English engine, and gives you a searchable PDF plus the extracted text to copy or download. It all happens in your browser.
What is a searchable PDF?
A scanned PDF is just images, so you can't select or search the text. OCR adds an invisible text layer on top of the page image, so the document looks identical but you can now select, copy and search the words. The picture of the page is unchanged, so a searchable PDF looks exactly like the scan it came from.
Why is some of the text wrong?
OCR accuracy depends on the scan: clear, high-contrast, straight pages read best; faint, skewed or handwritten pages read worse. This tool uses the English model, so other languages won't recognise well. The invisible text layer is aligned word-by-word, so it's great for search but not a perfect character-for-character copy.
Is my PDF uploaded anywhere?
No — the PDF stays on your device. The recognition engine (tesseract.js) and the English model run entirely in your browser — nothing is uploaded. Closing the tab clears everything.
Last updated: