What this PDF to text converter does
Copying text straight out of a PDF viewer often goes wrong: lines break in the middle of sentences, two columns get mixed line by line, and words split with a hyphen stay split. This tool reads every piece of text on each page together with its position and size, rebuilds the lines and paragraphs, puts them in reading order and gives you clean plain text. Copy all of it, copy one page, or download it as a TXT file that opens in any editor.
All of it happens in your browser with pdf.js, the open source PDF library from Mozilla. The file is never uploaded, so it is safe for private documents, and there is no size limit beyond what your device can handle. A progress bar shows how far a long PDF is, and you can stop at any time and keep the pages read so far.
Columns, tables and rotated pages
A PDF does not store paragraphs. It stores where each run of letters is drawn, and the order in the file is not always the order you read. So the converter looks at the layout instead:
- Columns are found from the empty gutter between them and read one after the other, left before right. A title or a figure caption running across both columns stays where it is, between the parts of the columns it separates.
- Tables with short cells keep each row on one line, with a tab between the cells. Labels and their values on the same line, like "Invoice date" and the date, stay together the same way.
- Rotated pages, landscape pages and text turned sideways are read in their own direction, so a page stored sideways or upside down in the PDF still gives normal lines.
- Doubled text drawn twice on the same spot, which makes letters look bolder, is only taken once.
The options
Join lines into paragraphs turns the lines of a paragraph into one flowing line, which is what you want when you paste into a document, an email or a translation tool. Short lines that stand on their own, like the lines of an address, a heading or a list item, keep their line break. Switch it off to keep every line break exactly as on the page, for poems, code or anything where lines matter.
Remove hyphens at line ends joins words that were split across two lines, so "infor-" and "mation" become "information" again. It only does that when the next line goes on in lowercase. Soft hyphens, the invisible break points some programs put into words, are always removed.
Page separators put a line like "--- Page 3 ---" before each page, so you can still find where a quote came from. Switch them off for one continuous text.
Pages limits the converter to the pages you need: type 5 for one page, 2-4 for a range, or 1-3, 8, 10- for a mix, where 10- means page 10 to the end. Pages already read are not read again when you change the range.
Scanned PDFs and pages without text
Some PDFs have no text at all, only pictures of pages. That is the case for most scans, photos of documents and faxes. When a page has no text, the tool tells you which ones, for example "No text found on pages 3 to 7", instead of quietly leaving gaps. For those pages, use image to text: it accepts PDFs too and recognizes the letters with OCR, also in your browser. If a PDF does have text but it comes out as random symbols, its fonts do not give their letters away, and OCR is the way to go as well.
Password-protected PDFs
If a PDF needs a password to open, a password box appears. Type the password and press Unlock. The password goes only to the PDF reader running in this browser tab. It is not stored, not written into the page address and not sent anywhere, and Clear forgets the unlocked file as well. A wrong password just asks again.
Arabic, Hebrew, Chinese, Japanese and Korean
Right-to-left text such as Arabic and Hebrew is put back in reading order, including numbers and Latin words inside a line. Chinese and Japanese lines are joined without inserting spaces, and vertical text is read top to bottom with the columns from right to left. The word count follows the rules of your browser for each language, so a Chinese sentence counts its words, not just its characters.
Tips for the best result
- To edit the text with its headings and lists in a word processor, PDF to Word may suit you better. To keep the look of a page, use PDF to JPG.
- For a very long PDF, set the pages you need first, so you do not wait for the whole file.
- To check the length of the result against a limit, paste it into the word counter.
How to convert a PDF to text
- 1Drop the PDF on the box or pick it from your files. If it asks for a password, type it in.
- 2Choose the pages and the options. The text updates right away.
- 3Press Copy all text, switch to One page to copy a single page, or download everything as a TXT file.
Frequently asked questions
Is my PDF uploaded?
No. Your browser opens the PDF and reads the text on your own device. Nothing is sent to a server, so contracts, bank statements and medical letters stay private.
Why does it say no text was found on some pages?
Those pages hold no text the PDF can hand over. Usually they are scans or photos of paper, where every page is one picture, and sometimes they are simply blank. Open the same PDF in the free image to text tool, which reads scanned pages with OCR (optical character recognition) in your browser.
Can I extract text from a password-protected PDF?
Yes, if you know the password. When a PDF needs a password to open, the converter asks for it. The password is only handed to the PDF reader in this browser tab to unlock the file. It is not saved, not put in the address bar and not sent anywhere.
Does it keep columns and tables in the right order?
It tries hard. The converter reads a page column by column, keeps a title or caption that runs across the columns in its place, and keeps each row of a table with short cells on one line, with a tab between the cells. Very unusual layouts, such as text boxes scattered across a poster, can still come out in an odd order.
Does it work with Arabic, Hebrew, Chinese, Japanese and Korean?
Yes. Right-to-left lines come out in reading order, Chinese and Japanese lines are joined without extra spaces, and vertical columns are read from right to left. If a PDF comes out as random symbols in any language, its fonts do not give their letters away; read those pages with image to text instead.
Why are some words joined or split in odd places?
Words split with a hyphen at the end of a line are joined again when Remove hyphens is on. A word only loses its hyphen when the next line goes on in lowercase, so names like Coca-Cola keep theirs, but a lowercase compound broken at its own hyphen, like well-known, comes out as wellknown. Switch the option off to keep every hyphen.