OCR PDF
Extract text from scanned PDFs — get a .txt file or searchable PDF.
Drop your scanned PDF here
or click to choose a file
Your file never leaves your browser — processed locally, nothing uploaded
Extract text from scanned PDFs — get a .txt file or searchable PDF.
A scanned PDF stores text as pixels, not characters — you can see the words but you can't select, copy, or search them. PDFree unlocks the text using OCR and gives you two outputs: a plain .txt file you can paste into any editor, and a searchable PDF where Ctrl+F and text selection work directly.
No upload. No account. No page limit on desktop. Your document stays on your device the entire time.
All extracted text with page separators. Paste directly into Word, Google Docs, Excel, Notion, or any text field. Best for editing or reusing content.
Original scan preserved exactly. Ctrl+F and copy work in any PDF reader. Best for archiving, sharing, or working inside a PDF viewer.
Enable "Also download .txt copy" in the OCR options to download both files at the same time.
PDFree immediately checks if the file is image-only or already has a text layer. Text-layer PDFs are extracted instantly with no OCR download needed.
In the OCR options, tick Also download .txt copy. This gives you the raw text as a separate file alongside the searchable PDF.
18 languages supported. Select the one that matches your document — accuracy depends on this. Install the OCR engine if prompted (one-time 17 MB, cached forever after).
Both files download automatically. Open the .txt in any editor or paste directly. Open the searchable PDF in any PDF reader for Ctrl+F and text selection.
Accuracy depends on the scan quality. PDFree shows a confidence score per page after OCR — you can see exactly which pages extracted cleanly.
| Scan condition | Expected accuracy |
|---|---|
| 300 DPI, clean white background, standard font | Excellent (90%+) |
| 200 DPI, minor shadows or slight skew | Good (80–90%) |
| 150 DPI or colored background | Fair (60–80%) — review recommended |
| Below 150 DPI, heavy noise, handwriting | Poor — rescan at higher DPI for best results |
No. OCR runs entirely in your browser using Tesseract.js. Your PDF never touches any server. Open DevTools → Network and you will see zero uploads while processing.
Yes. For PDFs created digitally (from Word, Google Docs, or any app), PDFree detects the existing text layer and extracts it instantly — no OCR engine or download needed. The result is the same .txt file and searchable PDF.
18 languages: English, French, German, Spanish, Italian, Portuguese, Russian, Uzbek, Dutch, Polish, Turkish, Arabic, Japanese, Chinese Simplified, Chinese Traditional, Korean, Hindi, and Thai. Select the correct language before running OCR for best accuracy.
PDFree can attempt to read password-protected PDFs without a password when the restriction is only on editing or copying (not opening). If the PDF requires a password to open, you will need to remove the password first using a tool with the owner password.