Scanned PDF to Word: How to Get Editable Text (Free, Step by Step)

The short answer

A scanned PDF contains pictures of pages, not text, so a normal PDF to Word converter has nothing to extract. Lazyblink's PDF to Word tool tells you this with the message "No selectable text found — this looks like a scanned PDF." To get editable text, turn each page into an image with PDF to JPG, read the text from each image with Image to Text (OCR), then paste the text into Word or Google Docs. All three steps run in your browser.

How to tell if your PDF is scanned

Open the PDF in any viewer and try to select a single word with your mouse or finger.

• You can highlight individual words: the PDF has a text layer. Use the PDF to Word tool directly.
• The whole page highlights as one block, or nothing highlights: the page is an image. You need OCR (optical character recognition).
• Search (Ctrl+F) finds nothing, even for words you can clearly see: also a scanned page.

Phone "scanner" apps, office photocopiers and fax-to-PDF services usually produce image-only PDFs unless they offer an OCR option.

Some PDFs are mixed: typed pages with a few scanned pages (for example, a signed page added at the end). The PDF to Word tool extracts text from the typed pages and leaves the scanned pages empty, so check the output against the original.

Step 1: Turn the PDF pages into images

Open the PDF to JPG tool.

Drop the scanned PDF onto the page.

The tool renders every page at twice its normal display size and saves each as a JPG, which gives OCR more detail to work with.

Download the pages you need, or use the option to download all. Files are named after your PDF with _page1, _page2 and so on.

If the PDF is password-protected, it will not open until you remove the password with the software you normally use to open it.

Step 2: Read the text with OCR

Open the Image to Text (OCR) tool.

Choose the Text language before uploading. Options include English, Hindi, Bengali, Tamil, Spanish, French, German, Portuguese, Russian, Arabic, Chinese (Simplified) and Japanese. The language data downloads once, then OCR runs on your device.

Upload the first page image.

Wait for the progress bar. When it finishes you see the extracted text and a confidence percentage.

Click Copy text, or Download .txt to save it.

Click New image and repeat for the next page.

The tool processes one image at a time, so a 10-page document means 10 uploads. For long documents, keep a blank Word file open and paste each page's text as you go, with a page heading so you can match it to the original.

The confidence percentage is the OCR engine's own estimate. A low number is a warning to proofread carefully, but even a high number does not guarantee every character is right.

Step 3: Paste into Word and tidy up

Paste the text into Word, Google Docs or any editor.

Use Paste as plain text (in Word, Paste Options > Keep Text Only) so the formatting stays consistent.

Fix line breaks. OCR keeps the line breaks from the scanned page, so sentences are broken mid-line. The guide How to Remove Line Breaks From Text shows how to join them quickly.

Re-create headings, bold text and lists with Word's styles.

Proofread names, numbers, dates and amounts against the original scan. These are the most important errors to catch.

What this workflow keeps and what it loses

Element	Result
Body text	Extracted as plain text
Line breaks	Kept as on the page, often needing clean-up
Fonts, bold, italics, colours	Not kept
Tables	Text extracted, but cells may run together or lose alignment
Images, logos, signatures	Not included in the text
Handwriting	Usually poor or unreadable
Page layout and columns	Not kept; two-column pages may mix lines from both columns

If the original layout matters (a form, a certificate), keep the PDF as the official copy and use the extracted text only for editing or searching.

How to get better OCR results

Most OCR errors come from the scan, not the software:

• Scan at a decent resolution. If you control the scanner, 300 dpi is a common choice for text documents.
• Keep pages straight. Tilted lines cause misreads. Re-scan rather than trying to fix a badly skewed page.
• Good contrast. Dark text on a clean white background works best. Shadows from phone photos, coloured paper and faded print reduce accuracy.
• Pick the right language. English OCR on a Hindi page produces nonsense. For Hindi documents, see How to Extract Hindi Text From an Image.
• One language per run. If a page mixes two scripts, run it once in each language and combine the parts.
• Crop out clutter. Stamps, logos and margin notes can confuse the engine. Cropping the image to the text area helps.

Troubleshooting

PDF to Word says "No selectable text found"

That is the scanned-PDF case. Follow the three steps above.

PDF to Word says the PDF is password-protected

The tool cannot read encrypted files. Open the PDF in a viewer where you know the password, and save or print a copy without the password before converting.

The OCR text is full of random characters

Check the language setting first. Then look at the image: if the page is blurry, very small or at an angle, re-scan or re-photograph it.

Numbers come out wrong

OCR often confuses 0 and O, 1 and l, 5 and S, and 8 and B. Always check amounts, account numbers and dates by eye.

Privacy

The PDF to JPG, Image to Text and PDF to Word tools all process files inside your browser. Your scanned documents are not uploaded to a server, which is important for bank statements, ID documents, contracts and medical letters.

For text-based PDFs and what the Word output looks like, see PDF to Word Conversion: Complete Guide with Tips.