How to Extract Text from a Scanned PDF (OCR, Free, in Your Browser)

A scanned document — an old contract, a photographed form, a PDF made from a stack of paper on a scanner — is really just a picture of text, not actual text. You can't select it, search it, or copy a sentence out of it. Getting real, usable text out of a scan needs OCR (optical character recognition), and most free OCR tools do that by uploading your document to a server.

OCR That Runs Entirely on Your Device

QR Ivify's PDF OCR uses Tesseract, a mature open-source OCR engine, compiled to WebAssembly and run entirely in your browser. Every page is rendered and read directly on your device — nothing is uploaded, which matters if the document has anything you'd rather not send to a third-party server.

Step by Step

1. Open PDF OCR.
2. Drop in a scanned or image-only PDF.
3. Choose which language(s) to recognize — English, Vietnamese, or both.
4. Click "Extract text." Each page renders and gets OCR'd in your browser — this can take a while on longer documents, since there's no server doing the work in the background.
5. Copy the result, or download it as a .txt file.

Getting the Best Results

OCR accuracy depends heavily on the scan itself. A clean, high-resolution, reasonably straight scan reads far better than a blurry photo, a skewed page, or a low-quality fax. If the extracted text has noticeable gaps or errors, the usual culprit is scan quality rather than the OCR engine — always give the result a read-through before relying on it for anything important.

Keep in mind this pulls out plain text in reading order — it doesn't preserve the original layout, columns, or tables from the scanned page.

Frequently Asked Questions

Does this upload my PDF anywhere? No. Every page is rendered and OCR'd entirely in your browser using Tesseract.js — nothing is sent to a server.

What languages are supported? English and Vietnamese, individually or together. Select whichever language(s) match your document before extracting.

Why is the extracted text missing words or has typos? OCR accuracy depends on scan quality — low resolution, skewed pages, or unusual fonts all reduce accuracy. Always review extracted text before relying on it.

Does this preserve the original page layout? No — the output is plain text in reading order, without columns, tables, or other formatting from the original page.

Is there a file size limit? Yes, 40 MB.

Try It

Have a scanned PDF you need real, copyable text from? QR Ivify's free PDF OCR reads it entirely in your browser, no upload required. Need to convert the result into a document afterward? See PDF to Word.