PDF to text: copy the words in a scanned PDF
Got a scanned contract, handout or report you can't select text from? Choose a PDF and press Recognize text. Pages without a text layer use OCR automatically, and big files show a warning so you can process just the first pages.
How to use
- Choose a PDF and the language of the text in it.
- Pick the pages (optional) and press Recognize text.
- Edit the result, then copy it or download a .txt file.
PDF to text: copy the words in a scanned PDF
Each page is checked first: if it has a text layer, that text is used as is (fast and exact). Only pages without one are drawn to an image and recognized with the Tesseract engine in your browser, one page at a time so memory stays low. Nothing is uploaded.
Long PDFs can be slow, especially on a phone, so you will see a warning for large files and can process only the first pages or cancel at any time. Scanned text is approximate; proofread the result.
Frequently asked questions
How do I turn a scanned PDF into copyable text?
Choose the PDF and the language of its text, then press Recognize text. Pages without a text layer are recognized with OCR one at a time, and you can copy the result or download a .txt.
Why are some pages marked “text layer”?
Those pages already contain real text, so the tool read it directly without running OCR. That is faster and exact.
Is my PDF uploaded?
No. Reading and recognition happen inside your browser, so contracts and reports stay on your device.
How many pages can I recognize?
On a phone, more than 10 pages or 15 MB shows a warning (double on a computer) and there is a hard limit. You can also enter a page range such as 1-5.