Tesseract.js runs entirely in your browser using WASM. Expect 30–120 seconds per page depending on your device.
Your everyday toolkit
PDF ToolsPDF OCR
Extract text from scanned PDFs or image-based PDFs using OCR (Optical Character Recognition). Uses Tesseract.js which runs entirely in your browser.
A little less effort
PDF OCR
Extract text from scanned PDFs with OCR
Drop a scanned PDF to extract text
or click to browse · PDF only · English text
Using PDF OCR well
Recognize English text in scanned PDF pages locally for copying or exporting as plain text.
Example to try
Try one clear, upright scanned page first. Compare names, dates and numbers with the image before processing a longer document.
What to check before relying on the result
Recognition is not exact and does not create a searchable PDF. The OCR runtime and English language data download when needed. Low-resolution scans, handwriting and other languages may perform poorly.
How to use PDF OCR
A scanned document
Image of a pageYOU’RE EXPLORING
Upload a scanned PDF
A few steps.
A whole new result.
Frequently Asked Questions
The useful details, before you get started.
No. OCR runs entirely in your browser. Nothing is sent to a server.
This tool recognizes English text. Multi-language support would require loading additional Tesseract language packs.
OCR quality depends on image resolution and clarity. Very low-resolution or skewed scans will produce poor results.
Choose any question to see the details.
