Extract text Pull the text layer out as .txt · takes PDF

The text layer of every page as a .txt file. Keep the line breaks, or reflow into paragraphs with hyphenated words rejoined.

0 bytes uploaded · your files stay on this device

How it works

Lines are reconstructed. A PDF stores glyphs at coordinates, not lines. Runs are grouped by baseline, read left to right, and a gap wider than the type around it becomes a space.

Scans have no text. A page that is an image of text produces nothing here, and a document with no text at all is reported as a scan rather than returned as an empty file. Making a scan searchable needs OCR, which this toolkit does not do.

Page markers. Optional "--- page N ---" lines carry the real source page number.

Related tools

Open your browser's network tab and run this tool: nothing is sent anywhere. The counter above is live, fed by every way a page can upload, and disconnecting from the internet does not stop the tool working once the page has loaded.