Extract Text from PDF

Extract text from scanned or image-based PDFs.

or drag and drop your PDF here

All processing happens locally in your browser. No files leave your device.

How to extract text from a scanned PDF

Drop your PDF into the tool and select the document's primary language. Click Extract Text (OCR). The tool renders each page as an image, then runs Tesseract.js | an open-source OCR engine | on each one. The extracted text appears in the output panel, separated by page, and can be copied to the clipboard or downloaded as a .txt file.

What is OCR and when do you need it?

OCR (Optical Character Recognition) converts images of text into machine-readable characters. Scanned documents, photos of printed text, and PDFs exported from image-only sources contain text as pixels | not as selectable, searchable characters. OCR makes that text usable: you can copy it, search it, paste it into other documents, or feed it into analysis tools. The language model download (~10 MB on first use) enables accurate recognition for the selected language.

Frequently asked questions

How accurate is the OCR?

Accuracy depends on scan quality and font clarity. Clean scans of standard printed text achieve very high accuracy. Handwriting, unusual fonts, low-resolution scans, or pages with complex backgrounds will reduce accuracy.

Why does it take a while to process?

OCR is computationally intensive. On first use, the language model (~10 MB) must be downloaded. After that, each page takes 10–60 seconds depending on complexity and your device's speed.

What languages are supported?

English, French, German, Spanish, Portuguese, and Simplified Chinese. Each uses a different Tesseract language model downloaded on demand.

Is my PDF sent to a server?

No. Rendering and OCR both happen locally in your browser using PDF.js and Tesseract.js. Your document is never uploaded anywhere.

Page share preview

More free image tools