PDFVela — pdfvela.com

How to copy text from a scanned PDF

Why the text will not select, and how to get it out without retyping

By · Updated October 26, 2026

Why you cannot copy the text

When a paper document goes through a scanner or a phone camera, the result is a photo of the page. It may be saved as a PDF, but to the computer each page is just coloured dots — there are no letters inside. That is why the text will not highlight, search finds nothing, and copy-paste gives you an empty line or the whole page as a picture.

To get the words out, the computer first has to read them from the image. This is called OCR (optical character recognition).

Step 1: recognise the text

  1. Open OCR PDF and choose the scanned file.
  2. Pick the language of the document. Choose English + Turkish for mixed documents — the Turkish model also knows letters like ş and ğ.
  3. Keep Skip pages that already contain text switched on if only some pages are scans.
  4. Click Recognise text and download the result.

The new PDF looks the same as before, but now you can select words, search with Ctrl+F (⌘F on a Mac) and copy paragraphs like in any normal document.

Step 2: get the text where you need it

  • A few sentences: open the recognised PDF in any PDF viewer, select and copy.
  • All the text at once: put the recognised file through PDF to text and copy it in one click, or download it as a .txt file.
  • An editable document: convert it with PDF to Word when you want to fix wording, change numbers or reuse the layout.

Getting better results

OCR can only read what it can see. A few minutes spent on the scan saves a lot of corrections later:

  • Scan straight and flat. Pages photographed at an angle are harder to read; our guide to scanning with your phone covers light, angle and shadows. A PDF that is already scanned crooked can be levelled with Deskew PDF.
  • Turn sideways pages upright with Rotate PDF before recognition.
  • Cut away dark edges and fingers with Crop PDF; they can turn into stray characters.
  • Use enough resolution. Around 300 dpi is ideal for normal print. Very small print benefits from a closer photo.
  • Check the important parts. Read through names, dates, IBANs and amounts — a 0 read as an O is easy to miss.

When to type it out instead

If the page is handwritten, very faded or only a line or two long, typing it yourself can be quicker than correcting the recognised text. For everything else — contracts, letters, invoices, book pages — OCR saves you from retyping whole pages.

Ready to try it? It is free and your file stays on your device.

Open OCR PDF

Copying text from scans

How do I know if my PDF is scanned?

Try to select a single word. If the whole page gets highlighted as one block, or nothing is selected at all, the page is an image. Searching for a word you can clearly see and getting no results is another sure sign.

How accurate is the recognised text?

On a clean, straight, well-lit page it is usually close to perfect. Accuracy drops with blurry photos, tiny print, handwriting, coloured backgrounds and pages scanned at an angle, so always read through names, numbers and amounts before you use them.

Can OCR read handwriting?

Only neat block letters, and even then with mistakes. OCR is built for printed text. Handwritten notes are usually faster to type out yourself.

Does the recognition change how my document looks?

No. The text is added as an invisible layer behind each page, so the pages look exactly the same. You only notice the difference when you search, select or copy.

Is my document uploaded for recognition?

No. The recognition engine runs in your own browser, so the scanned pages never leave your device.