PDF to Text

About PDF to Text

This pulls the plain text out of a PDF so you can reuse it — paste it into a document or email, quote from a report, feed it into a translator or notes app, or strip away formatting to get just the words. It reads the text as it's stored in the file, page by page, and hands you the result to copy or download as a .txt.

Extraction runs on your device, so even confidential documents stay private.

How it works

The PDF is opened with pdf.js in your browser. For each page, the tool reads the page's text layer — the actual character and positioning data the PDF stores — and appends it, separating pages with a blank line. It streams the text out as it goes, so you see it fill in page by page, and shows a running "page X of N" status. Nothing is retyped or guessed; it surfaces the text that's genuinely in the file.

Assumptions and defaults

• It reads the embedded text layer, so results depend on how the PDF was made. A PDF exported from Word, a browser or most software has a clean text layer; a scanned or image-only PDF has none.
• Pages are output in document order, separated by a blank line.
• The download is a plain-text .txt named after the source PDF.
• If no selectable text is found, the tool tells you explicitly that the PDF looks scanned and would need OCR.

Limitations

• No OCR. This reads existing text; it does not recognise text inside scanned images or photos. If the PDF is a scan, you'll get an empty result and a note that OCR is needed — this tool can't do that step.
• Layout is not fully reconstructed. Text comes out in the order the PDF stores it, which is usually correct for single-column documents but can interleave or misorder columns, tables, headers/footers and text boxes in complex layouts.
• Formatting is dropped — you get plain text, not fonts, bold/italic, headings, or table structure.
• Encrypted or corrupted PDFs can't be read and will show an error.

Privacy

Text is extracted with pdf.js entirely in your browser. Your PDF and its text never leave your device — nothing is uploaded or stored.