PDF to Text (OCR)
Get editable text from any PDF, scan or screenshot: real text is copied as it is, and OCR reads only the pages that need it.
- Files never leave your device
- 100% free
- No sign-up
- No watermark
- Works offline after first load
or drag & drop scanned PDFs, photos or screenshots here
How to extract text from a PDF or scan
- Select or drop PDFs, scans, photos or screenshots — several at once.
- Pick the language (and a second one for mixed text). Smart mode uses the PDF’s own text and runs OCR only where it’s needed.
- Click Extract text, then copy it from the preview or download the .txt file.
About PDF to Text (OCR)
Get clean, copyable text out of PDFs, scanned documents, photos and screenshots. Pages that already contain text are read directly and exactly; scanned pages and pictures are read with OCR in any of 28 languages, two at once for mixed text such as Hindi and English.
Wrapped lines are joined into paragraphs, hyphenated words are mended and each page is marked, so the result is ready to paste into an email or document. Preview and copy the text right on the page or download it as .txt — everything runs in your browser, so nothing is uploaded.
Good for
- Copying text from a scanned letter or notice
- Getting the words out of a photo of a document or whiteboard
- Pulling all the text out of a long PDF report
Tip: Need tables and formatting too? PDF to Word rebuilds them as an editable document.
Frequently asked questions
What’s the difference from OCR PDF?
PDF to Text gives you the text itself — to copy, paste or save as a .txt file. OCR PDF keeps your PDF and makes it searchable. To get an editable Word document with tables, use PDF to Word.
Does it work with normal (non-scanned) PDFs?
Yes. In Smart mode the text already in a PDF is used directly — instantly and exactly — and OCR runs only on scanned pages and pictures. Choose “OCR every page” if a PDF’s text copies out garbled.
Can I extract text from a photo or screenshot?
Yes. Add JPG, PNG, WebP, GIF or BMP images alongside PDFs; each is read with OCR. Turn sideways photos upright in the Pages view first.
Which languages are supported?
28 languages including English, Hindi, Marathi, Bengali, Tamil, Telugu, Gujarati, Punjabi, Urdu, Arabic, Chinese, Japanese, Korean, Russian and the main European languages — two at once for mixed text.
Is the layout kept?
Paragraphs, headings and list items are kept on their own lines, wrapped lines are joined into paragraphs and words split by hyphens are mended. Untick these under Text layout to keep every line exactly as on the page.
Can I process several files at once?
Yes. Each file gets its own .txt (download them separately or as a ZIP), or tick “One text file for all files”.
Do my files stay on my device?
Yes. Reading and OCR run in your browser. Only the OCR engine and language data are downloaded the first time you use them.
Why does the recognised text have mistakes?
OCR is only as good as the scan. Blurry, tilted or very small text lowers accuracy — scan at 300 DPI if you can, pick the right language, and straighten crooked pages with Deskew PDF first.