Converting a PDF to text gives you just the words, without layout, fonts or pictures, in a .txt file that opens anywhere. It is the quickest way to search a long document, quote from it, or feed it to another program. Here it runs in your browser, so your documents are never uploaded.
Extract text from several PDFs
Add a whole folder of reports, articles or statements. Each PDF becomes a .txt file with the same name, and they are saved together in a ZIP. The text follows the reading order of each page, page after page, in UTF-8, so accents and non-Latin alphabets come through.
Scanned PDFs need OCR
This tool reads the text that is stored inside a PDF. A scan or a photo saved as PDF only contains pictures of letters, so the result is empty. Quick test: if you cannot select a word in your PDF viewer, use PDF OCR, which recognizes the letters and also gives you a text file.
Why text will not copy from some PDFs
Copying from a PDF viewer often breaks lines in the middle of sentences, mixes columns or picks up headers and page numbers. Some PDFs block copying with a permissions password; those can be opened with Unlock PDF first. And in a few files the letters are drawn with fonts that have no proper character map, so the extracted text looks like gibberish; running OCR on such a file gives readable text.
How it works
Text is extracted with MuPDF, a PDF engine running in your browser. It reads the characters stored in each page together with their positions, puts them into reading order and writes them as plain text. The first time, the engine (about 39 MB) is downloaded and kept by your browser. Your files are never sent anywhere.
What people use it for, and when it will not work
Searching and quoting long reports, collecting text for research or notes, pasting a contract into an email, importing documents into a translation or AI tool, and keeping plain-text archives.
- Formatting is not kept: no bold, tables or pictures. For that, use PDF to Word.
- Scans give no text without OCR.
- Complex layouts such as tables and multi-column pages may come out in an unexpected order.