Extract the text already inside a PDF
PDF to Text reads embedded text objects and writes a TXT file with a separator for each page. It is useful when you need to quote passages, search a document in a text editor, or compare wording between two versions. It does not turn a photograph of a page into recognized text.
How to use PDF to Text
- Select Choose one PDF. The file must be unencrypted and no larger than 25 MiB, with at most 200 pages.
- Select Extract text. The PDF.js worker and supporting font assets are served by this site; your selected document is processed in the browser.
- Read the Extracted result and select Download text (.txt) to save the TXT file.
- Compare a few lines with the original, especially if the PDF contains columns, tables, footnotes or unusual fonts.
For a reproducible check, download the two-page sample PDF. Select it and extract the text. The result should contain --- Page 1 --- and --- Page 2 ---, with the embedded page text beneath those separators. A page without extractable text is marked [No embedded text found on this page]; that marker is not text from your document.
Why text order or spacing can look different
PDF stores drawing instructions rather than an ordinary document's paragraph structure. This tool uses the sequence and line-break hints returned by PDF.js. A two-column page may be interleaved, a table can lose its columns, and fonts with incomplete character mappings can produce incorrect letters. The output is an extraction, not a faithful reconstruction of the original layout. Review it before using it as a quotation or a source of numeric data.
Scanned pages and extraction limits
A PDF can display readable words while containing only an image. Such pages need OCR. Use PDF to Images to render pages, then Image to Text for the existing English OCR workflow. This is a separate step and can introduce recognition errors.
The tool stops above two million extracted characters rather than silently truncating the result. For an oversized document, use Extract PDF Pages to make a smaller input. Encrypted, damaged or unsupported PDFs may fail to load. Try exporting an unencrypted copy from the original application if you have permission to do so.
Compare or reuse the result
Use Text Compare for differences between extracted versions. Differences in spacing or extraction order are not necessarily changes to the original document. No document content is placed in the URL or uploaded by this tool. Closing the page discards the in-memory result; keep a downloaded copy if needed.





