PDF to Text
Save the words of a PDF as a plain text file, ready to paste into an email, search, count or translate. Fonts, layout and pictures stay behind; the text follows the page from top to bottom.
Add one PDF file to convert to text.
or drop a file here
Your PDF is handled inside this browser tab and is never uploaded.
How to use it
- Add the PDF: press ‘Choose a file’ or drag the file onto the page.
- To extract from part of the document, choose ‘Page range’ under ‘Pages to convert’ and list pages like 2, 5-7.
- If any pages are scans, select the languages of the document under ‘Text recognition for scanned pages’. Pages that already contain text are read directly.
- Press ‘Convert to text’, then ‘Download’ to save the .txt file under the PDF's name.
- Open the file in any text editor, or paste its contents wherever you need them.
What it does
Paragraphs, not broken lines
Lines that sit close together at the same size are joined into one paragraph, words split with a hyphen at the end of a line are rejoined, and paragraphs are separated by a blank line. A bullet or a numbered item starts a paragraph of its own.
Tables as tab-separated rows
Each table row becomes one line with a tab between the cells. Paste it into a spreadsheet and every piece lands in its own cell.
Page markers
When you convert more than one page, each page begins with a marker line (— Page 3 —, for instance) that carries the page's number in the PDF, so a quotation can be traced to its page. A single page gets no marker.
Scans read with OCR
A page with no text layer is read by OCR (text recognition), using the languages you choose: English, Korean, Japanese, German, French, Spanish and Portuguese are offered. The tool tells you how many pages were read this way, because those are the ones to proofread.
UTF-8 that opens correctly
The file is UTF-8 and starts with a byte-order mark, which lets older Windows programs recognize accented letters, Korean and Japanese right away. In other programs, pick UTF-8 if you are asked.
Nothing leaves your device
Contracts, letters and records are read in your browser. The PDF is not uploaded, and it is not changed.
Good to know
Plain text is the plainest format there is: every program on every device opens it. People pull the words out of a PDF to quote a passage in an email, run a translation, count words against a limit, compare two versions of a contract, build a searchable archive, or hand the content to software that cannot read PDFs.
What comes out depends on how the PDF was made. A file exported from a word processor has a clean text layer. A PDF printed from a web page or an old system can have fonts with no map back to real letters, and the output then shows odd symbols; PDF OCR can read such a page from its picture.
Pages set in side-by-side columns are read straight across, so the two columns can come out paired on one row, separated by a tab. Headings, bold, italics, fonts and pictures are not recorded; a heading is just another line. PDF to Word keeps headings and tables as editable text, and PDF OCR shows the text beside each page so you can correct it and save it.
Questions and answers
Can I copy the text without saving a file?
This tool saves a .txt file rather than showing the text on the page. PDF OCR shows each page next to its text, with a button that copies everything.
Why does the text contain strange symbols?
The PDF's font has no usable map from its shapes back to letters, which is common in files printed to PDF from a browser. Open the PDF in PDF OCR and ask it to recognize the page anyway, which reads the picture instead of the broken text layer.
Why are my two-column pages mixed up?
Lines are read straight across the page, so text set in side-by-side columns comes out with the left and right pieces on one row, separated by a tab. Pasted into a spreadsheet, the two columns land in separate cells, and you can copy one after the other.
Can it read a scanned PDF?
Yes, once you select the document's language under ‘Text recognition for scanned pages’. A few characters may be misread, so give names and numbers a quick check.
Which encoding does the file use?
UTF-8, with a byte-order mark at the start, which Windows Notepad recognizes. If another program shows odd characters, reopen the file and choose UTF-8.
Are headings, bold text and pictures kept?
No. The file holds text only, and a heading appears as an ordinary line. PDF to Word keeps headings and tables as editable text, and PDF to JPG saves pages as pictures.