PDF to Text
The words out, the document stays
…
- 1
Drop the PDF in
It is read from disk and never sent anywhere, which for a contract, a payslip or a medical letter is the whole point.
- 2
It reads the pages
A fifteen page paper takes about a tenth of a second. Multiple columns are put back in reading order rather than left interleaved, which is what makes a naive copy and paste out of a PDF unreadable.
- 3
Copy it or download it
Plain text, as a .txt file. If you want the headings and tables kept, PDF to Markdown is the same read with the structure left in.
FAQ
- Is my document uploaded?
- No. The reader is a small piece of code that runs on this page, on your machine, and no request carries the document anywhere. Open the network panel and drop a file in if you would like to watch.
- Why does nothing come out of my PDF?
- Because it is almost certainly a scan. A scanned page is a photograph of words, and a photograph has no text in it to take out, however clearly you can read it. The page will say so rather than hand you an empty file, and Image to Text is the tool for that case: it looks at the picture and works out the letters.
- Why is the copy and paste out of my PDF viewer worse than this?
- Mostly columns. A PDF stores where each run of text sits on the page and not what order to read it in, so a two column page copied naively comes out as the first line of the left column, then the first line of the right, and so on. This works out the columns and puts them back in order.
- Some of it came out as gibberish. Why?
- The document embeds its fonts without saying which character each shape stands for. The letters are drawings at that point, and nothing can recover the text from the file alone. Running it through Image to Text sidesteps the problem by reading the shapes instead.
- Is there a size limit?
- Not a fixed one. It is your own machine doing the reading, and a few hundred pages is still quick, so the limit is whatever your browser will hold rather than a number we picked.