Extract text from a PDF
Drop in a document and its text appears below once every page has been read, ready to copy or download. It is read in this tab, so the file never leaves your machine.
How to use it#
- Drop a PDF onto the box above, or click it to choose one.
- Read or copy the text from the box. Untick Mark where each page starts if you want one unbroken text.
- Download it as a .txt file if you want to keep it.
What comes out, and what does not#
A PDF that was made from a word processor, a web page, or a spreadsheet holds its words as real text. This tool reads that text back out, line by line, with a blank line between paragraphs and a marker at the top of each page. Accents, symbols, and other alphabets come through as they are.
The layout does not come with it. Fonts, bold and italic, colours, and pictures are gone, and a table comes out as rows of values separated by spaces. The text is the part worth having when you need to quote it, search it, translate it, or paste it into something else.
Scanned documents have no text to extract#
A scan is a photograph of a page. It looks like text, but the file holds only pixels, and there are no letters in it for any tool to read out. Many PDFs from a scanner or a phone camera app are like this, and so are older documents scanned for an archive. Some scanning apps add a hidden text layer as they go, and those do come out.
When a page has no text, the tool says so and names the page rather than handing back an empty box. Getting words out of a scan takes OCR, which recognises the letters in the picture. The PDF OCR tool does that in your browser as well, and gives you a searchable copy of the file too. It downloads a recognition model of a few megabytes, which is why it is a separate page.
Why reading order can go wrong#
A PDF places each run of text at a point on the page, and the order it stores them in is up to whatever made the file. Most documents store them in reading order, and the text comes out the way you would read it. A two column layout usually comes out one column after the other, because that is the order it was written in.
Some do not. A document built from boxes laid out by hand, a form, or a poster can store its text in whatever order the boxes were drawn, and then lines come out in that order instead. Headers, footers, and page numbers are text too, so they appear on every page.
Frequently asked questions#
Is my PDF uploaded to read the text?
#
No. The document is opened by code running in this tab, and the text is pieced together here. Nothing is sent to a server, so a contract or a medical letter stays on your machine.
Why is some of the text missing or out of order?
#
Missing text is usually a scanned page, which holds a picture of the words rather than the words themselves, and the tool names those pages. Text out of order comes from a document that stores its text in a different order from the one it is read in, which some forms and designed layouts do.
Why can I not copy text from my PDF?
#
Usually for one of two reasons. The page may be a scan, a picture of words with no text behind it, and this tool names those pages; the PDF OCR tool reads the words in them. Or the author locked the file against copying, which greys out Copy in most readers, and this tool refuses it too. The file’s owner password lifts that lock in the PDF password remover, or the author can send you a copy without it.
How do I convert a PDF to a TXT file?
#
Drop the PDF in and press Download .txt. The file is plain text in UTF-8, which Notepad, TextEdit, and every editor open with accents and other alphabets intact. To paste the words somewhere instead, copy them from the box.
Can I get the text into Word or a spreadsheet?
#
Copy it from the box and paste it in. The words and paragraphs come across, but not the formatting, and a table arrives as lines of values separated by spaces, which a spreadsheet will need splitting into columns. For Word, the PDF to Word tool saves the same text as a .docx file, one paragraph at a time.
Can it read a scanned PDF?
#
Not on its own. A scan has no text layer, only a picture of the page, and turning that into words needs OCR. This tool tells you which pages are scans, and the PDF OCR tool reads them, still without uploading anything.
Does it work on a password protected or copy protected PDF?
#
Only once the password is off. A file that asks for a password cannot be read without it, so the tool says so. Remove the password with the PDF password remover, then bring the unlocked copy back here. A file that opens freely but whose author forbade copying is refused as well, because taking its text out is the copying the author ruled out. Its owner password lifts that lock in the same remover, or the author can send a copy without it.
What are the page markers for?
#
Each page starts with a line such as --- Page 3 ---, so a passage you find in the text can be found again in the PDF. Untick Mark where each page starts to get the text as one unbroken piece instead, with the pages run together.