Extract Text
Extract Text pulls the words out of a PDF and gives you a plain .txt file. It reads the document's text layer in your browser — and tells you plainly when there isn't one, which is what happens with a scan.
Your document stays on your device.
The PDF is parsed by pdf.js in this tab and the text file is generated locally. Contracts, invoices and reports are exactly the kind of document that should not be uploaded to read.
Learn how files are processed →- Input
- PDF with a text layer
- Output
- Plain .txt, pages in order
- Won't work on
- Scans — those need OCR
- Processing
- pdf.js in your browser
How to use Extract Text
- 01Upload
Add your PDF
Drop the file in. Nothing is uploaded.
- 02Customize
Check the preview
The tool shows how much text it found and what the first page looks like.
- 03Process
Download the text
Save it as a .txt file, with pages separated in order.
What is Extract Text?
Most PDFs contain a text layer: the actual characters, with a position and a font for each. That is why you can select a sentence in a PDF reader and copy it. Extract Text reads that layer for every page and writes it out as plain text.
It is the fastest way to get a quotable, searchable, editable version of a report, a paper, an invoice or a contract — particularly when the document is long and copying page by page is tedious.
The important caveat is scans. A photographed or scanned page has no text layer at all, and no tool can extract characters that were never stored. This one says so rather than handing you an empty file.
Supported formats
- Maximum file size —
- bounded by your device's memory
- Processed by —
- pdf.js, on your device
Limitations
Worth knowing before you start, rather than after.
- A scanned PDF has no text layer — it is a picture of a page — so there is nothing to extract. That needs OCR, which this tool does not do.
- Complex layouts such as multi-column pages and tables lose their arrangement. The words are all there; the columns are not.
Why PDF text comes out in the wrong order sometimes
A PDF does not store a document the way a word processor does. There are no paragraphs, no reading order and no columns — only instructions to draw particular glyphs at particular coordinates. The visual structure you see is an emergent property of where things were placed.
Extracting text therefore means reconstructing structure that was never recorded. This tool groups fragments that share a baseline into lines, then orders lines down the page and fragments across it. For ordinary single-column prose, that recovers the text almost perfectly.
Two layouts defeat it, and defeat every non-AI extractor for the same reason. Multi-column pages interleave: two columns side by side share baselines, so a naive reconstruction reads across both columns instead of down one. Tables lose their grid entirely, because a table in a PDF is usually just text positioned to look like a table, with the lines drawn separately.
If the output looks scrambled, that is what happened, and it is worth checking whether the original document exists in another format before trying to repair it by hand.
One more quirk: some PDFs embed fonts with non-standard character mappings, where the glyph drawn on screen and the character code stored underneath disagree. Those extract as nonsense — recognisable because the output is the right length and shape but the letters are wrong. Nothing can be done about that short of OCR.
Your file, and what the page does
Your file is never uploaded. Extract Text reads it into this browser tab and processes it with pdf.js, Mozilla's PDF renderer on your own device. There is no upload endpoint in the application and no copy on any server — closing the tab discards it.
What the page does send or fetch
- Ads on the page
- The site is funded by advertising, so the page loads Google AdSense. That carries the usual web basics — IP address, browser, referring page — as on any ad-supported site.
- An anonymous usage counter
- When Extract Text finishes we record that the tool ran, whether it succeeded and how long it took. No file data, no identifier, no cookie.
Your file is not on that list. How this works.
Frequently asked questions
Almost certainly because the PDF is a scan. A scanned document is a picture of a page — the words are pixels, not characters, so there is no text layer to read. Recognising words in an image requires OCR, which this tool does not do.
Open it in any PDF viewer and try to select a line of text with the cursor. If you cannot, or if selecting drags a rectangle over the whole page, it is an image.
A PDF stores each fragment of text at a position on the page; it does not store paragraphs, columns or tables as structures. The tool reconstructs lines from those positions, which works well for ordinary prose and poorly for multi-column layouts and tables — the words are all present, the arrangement is not.
No. The output is plain text. Formatting, fonts, images and colours are not carried over.
Only if it opens without a password. An encrypted document cannot be read until it is unlocked.
No. The PDF is parsed in this tab by pdf.js and the text file is generated locally.
Yes — Extract Text is completely free. There's no account, sign-up, watermark or limit.
No. It runs in any modern browser — Chrome, Edge, Safari, Firefox, Brave — with nothing to download or install.
Need the words out of a PDF?
Drop the file in and see how much text it actually contains.
Edit any media with a right-click
Add Dexta Studio to Chrome and open any image, video, audio or PDF straight into the right tool — free, nothing uploaded.