The Quick Answer
To copy text from a PDF without broken lines, stray page numbers and split words, extract the text layer instead of selecting it in a viewer. Open the PDF to Text Converter, add your file, leave the layout on Paragraphs, keep “Remove page numbers & repeated headers/footers” switched on, and copy the result or download it as a .txt file.
If the tool says a page has no selectable text, that page is a scan: a picture of paper. Pictures need OCR, which this guide also covers. Both tools run in your browser, so the PDF is never uploaded.
Why Copying From a PDF Viewer Goes Wrong
A PDF is built to look identical everywhere, not to be edited. It stores text as pieces placed at exact positions on a page, with no record of where a paragraph starts or ends. When you select and copy in a viewer, you get the side effects of that design:
- A line break at the end of every visual line, so each sentence arrives in fragments.
- Words split by hyphens where the original was hyphenated at the end of a line (“infor- mation”).
- Running headers, footers and page numbers mixed into the text every time you cross a page boundary.
- Columns and tables in the wrong order, because the file records where text is drawn, not the order a person reads it.
The converter rebuilds the text from those positioned pieces. It uses Mozilla’s PDF.js, a web standards-based library for parsing and rendering PDFs (checked on 2026-10-14), running in a background worker in your browser.
Step by Step With the PDF to Text Converter
- Add your PDFs. Choose one or more files or drop them on the page. Each file gets its own panel. Files up to 100 MB can be opened.
- Enter a password only if asked. If a PDF needs a password to open, the panel asks for it. The password is used only in that browser tab and is not saved or sent anywhere. PDFs that merely restrict copying or printing open without one.
- Choose the pages. Leave Pages empty for the whole document, or type a selection (see the cheat sheet below). The text and counts update as you type.
- Choose the layout. Paragraphs re-flows lines into normal paragraphs; “Lines as in the PDF” keeps every visual line.
- Set the options. Keep page numbers and repeated headers or footers removed for clean text, or add page markers (“--- Page 3 ---”) when you need to cite page numbers later.
- Copy or download. Check the word and character counts, edit if you like, then copy or download a .txt file. When you selected pages, the file name includes them. With several PDFs, “Copy all text” and “Download all as one .txt” combine every panel.
Page Selection Cheat Sheet
These are the actual results the converter gives for a 20-page PDF:
| You type | Pages extracted |
|---|---|
(empty) or all | 1–20 |
1-3, 5 | 1, 2, 3 and 5 |
8- | 8–20 |
-4 | 1–4 |
10-last or 10 to end | 10–20 |
odd | 1, 3, 5 … 19 |
2, 2, 1 | 1 and 2 (document order, each page once) |
25 | Error: the PDF has 20 pages, so page 25 doesn’t exist |
3-1 | Error: ranges run from low to high, like 1-3 |
Every page is read once when the file opens, so changing the selection or the layout afterwards is instant, even for long documents.
Paragraphs or Lines: Which Layout to Use
In Paragraphs mode, consecutive lines are joined unless something signals a break: a clearly larger vertical gap, a change of font size (a heading), a first-line indent, a bullet or numbered item, or a line that ends well short of the right edge. Words hyphenated at a line end are joined without a space, and a paragraph that continues onto the next page in lowercase is stitched back together. The hyphen itself is kept, because the converter cannot tell a split word from a real compound such as “well-known.”
| Document | Best layout |
|---|---|
| Reports, articles, papers, ebooks, letters | Paragraphs |
| Poetry, song sheets and scripts | Lines as in the PDF |
| Addresses, invoices and forms | Lines as in the PDF |
| Code listings and command examples | Lines as in the PDF |
| Simple tables | Lines as in the PDF, then tidy by hand |
| Slides exported to PDF | Lines, with page markers so each slide stays separate |
If you already copied text from a viewer and only need to fix the line breaks, Remove Line Breaks joins wrapped lines while keeping paragraph gaps. Our guide to removing line breaks explains its modes.
Worked Example: Removing Headers and Page Numbers
Imagine a four-page report where every page starts with a running header and ends with a page number. Copied from a viewer, page 2 looks like this:
Annual Report 2026 — page 2
Spending rose in two areas.
Travel costs fell.
- 2 -With the option on, the converter removes two kinds of line at the top or bottom of a page:
- Page numbers on their own line, such as 2, - 2 -, (2), Page 2, 2 of 4 or 2/4.
- Repeated first or last lines that appear on at least half of the pages (and at least three), compared with the digits ignored. “Annual Report 2026 — page 2” and “Annual Report 2026 — page 3” therefore count as the same header.
We ran exactly this case through the converter’s own clean-up function. It removed eight lines (four headers and four page numbers) and kept every body line. The panel tells you how many lines were left out. One caution: if every page of a form ends with the same real sentence, that line is removed too, so switch the option off when you need everything exactly as printed.
Quoting and Citing From a PDF
Students and researchers usually extract text to quote it, and quotations need page numbers. Three habits save time:
- Turn on page markers. Every page then starts with a line such as “--- Page 12 ---”, so each passage you copy can be traced back to its page.
- Cite the printed page number, not the marker. The markers count pages in file order, starting at 1 with the first page of the PDF. Books and journals often number their pages differently, for example with roman numerals for the front matter or with the page range of the journal issue. Look at the number printed on the page before you write the citation.
- Check every quotation against the page. A kept hyphen, a removed running header or a misread symbol can change a quotation, and quotations must be exact.
Scanned PDFs: How to Tell and What to Do
A scanned PDF contains images of pages, not text. Three quick signs: you cannot select individual words in your viewer, searching with Ctrl+F finds nothing, or the converter reports that a page has no selectable text. It lists the affected pages and still extracts the rest, which matters for documents that mix typed pages with a scanned signature page.
To get the text from scanned pages:
- Zoom in in your PDF viewer until the text is comfortably large. OCR reads big, sharp letters far better than small ones.
- Save each page as an image: many viewers can export pages as PNG or JPG, or take a screenshot (Win+Shift+S on Windows, Cmd+Shift+4 on a Mac).
- Add the images to Image to Text (OCR) and extract them together with “Extract all.” It reads English text and runs in your browser too.
- Check names and numbers against the scan. OCR is never perfect, and digits are where errors cost the most.
What Even a Good Extractor Cannot Fix
- Multi-column layouts and tables come out in the order the file stores them, which is not always reading order. Try Lines mode, or copy one column at a time from the viewer.
- Text drawn as outlines or images, common in design exports, logos and charts, looks like text but is not stored as text.
- Unusual font encodings in some older or generated PDFs produce symbols instead of letters. That is how the text is stored, and copying from any viewer gives the same result. OCR on a screenshot is the workaround.
- Form field entries and comments are not part of the page text and are not included.
Copy restrictions deserve a word too. The converter can read PDFs that restrict copying, because any PDF that opens without a password can be read by PDF software; the copy restriction is a flag that viewers choose to honor. Make sure you have the right to reuse the text, and cite it when you quote it.
What to Do With the Text Next
- Count it. The converter shows words and characters for your selection. For several documents at once, PDFs and Word files together, use the file counter in the Word Counter.
- Compare two versions. The Text Diff Checker opens PDF files directly and shows what changed between, say, two drafts of a contract.
- Translate it. Paragraphs mode gives whole sentences, which is what the Private Text Translator needs; it translates on your device in desktop Chrome.
Troubleshooting
“That password didn’t work”
Passwords are case-sensitive. Check the Caps Lock key, and make sure you are using the password that opens the file, not one that only controls editing.
Letters come out as symbols or boxes
The file uses a font encoding without a usable text mapping. Screenshot the page and use OCR.
Words run together or have extra spaces
The converter inserts a space where the gap between two pieces of text is wider than a fraction of the font size. PDFs with unusual letter spacing can produce a missing or extra space now and then. Fix those by hand, and use Remove Extra Spaces for doubled spaces.
The text is empty but the PDF clearly has words
The page is an image. Follow the scanned-PDF steps above.