How to Extract Text from a PDF Without Software (Free)
Summary: Most PDFs can be copied straight from your browser, and scanned ones need OCR. Here are the free methods, when each one works, and how to fix the usual problems.
You can extract text from a PDF for free without installing anything. If the PDF contains real text, open it in your browser and copy it, or upload it to a free online tool. If the PDF is a scan, you need OCR (optical character recognition), which both Google Docs and our free PDF to Text tool provide.
The right method depends on which kind of PDF you have, so start there.
Text-based and scanned PDFs
Two PDFs can look identical on screen and behave completely differently.
| Text-based PDF | Scanned or image PDF | |
|---|---|---|
| Usually comes from | Word, a website or an app that exported it | A scanner, a phone scanning app or photos |
| What the file holds | Characters, fonts and their positions | A picture of each page |
| Can you select a word? | Yes | Usually not |
| What you need | A browser or a text tool | OCR |
To find out which one you have, open the PDF and try to highlight a sentence. If individual words highlight, it's text-based. If nothing highlights, or the whole page is selected as a single block, it's a scan. Some files mix both, with typed pages next to scanned ones.
If you can see text but copying fails or pastes as nonsense, the cause may be something else. Our article on why you can't copy text from a PDF walks through each possibility.
Method 1: Copy it from your browser
This is the fastest option for a text-based PDF, and the file never leaves your computer.
- Open the PDF in your browser. Dragging the file onto an open browser window usually works.
- Select the text you want with your cursor, or press Ctrl+A (Cmd+A on a Mac) to select everything your viewer allows.
- Copy it, then paste it into your notes or document.
This works well for a paragraph or two. It gets clumsy on long documents, because headers, footers and page numbers come along with the text, and multi-column pages may paste in the wrong order.
Method 2: Use Google Docs OCR
If you already use Google Drive, this is a free way to get text from a scan. It needs a Google account.
- Upload the PDF to Google Drive.
- Right-click the file.
- Choose Open with, then Google Docs.
Google's help page sets a few limits. The file should be 2 MB or smaller, which rules out many multi-page scans. Google also says bold, italics, font size, font type and line breaks are likely to be kept, while lists, tables, columns, footnotes and endnotes are not likely to be detected. For best results, it recommends sharp images, even lighting and text at least 10 pixels high, with pages the right way up.
Method 3: Use a free online PDF to text tool
Our PDF to Text tool handles both kinds of PDF with no account.
- Open the tool.
- Drop your file in, or paste a link to a PDF that's publicly available.
- Wait for it to finish. A progress indicator shows the page it's working on.
- Edit the text if needed, then copy it or download it as .txt, .docx, .xlsx, .pdf, .html or .md.
Here's what happens to your file. A text-based PDF is read in your browser and isn't uploaded anywhere. If the tool finds almost no readable text, it treats the file as a scan and sends it to Google's Gemini API for OCR.
There are limits worth knowing. You get 10 extractions an hour, and 30 requests a day for anything that uses our server. A scan sent for OCR has to fit in a 5 MB request, which works out at a file of about 3.7 MB. If a scan is too big, or the AI service is unavailable, the tool reads up to the first 10 pages in your browser instead. That's slower and less accurate, but you still get a result.
For advice on scanning settings and preparing poor-quality documents, see our guide to OCR for scanned PDFs.
How to get text from an image inside a PDF
A text-based PDF can still hold pictures: a chart, a pasted screenshot, a photo of a sign. Tools that read a PDF's text layer skip those pictures, because there are no characters to read.
To get that text, take a screenshot of the image and upload it to our Image to Text tool. This also solves the mixed-file problem described below.
How to extract text from just one page
Our tool doesn't have a page picker, so there are two practical routes.
- Use the full result. For a text-based PDF, run the whole file. Each page is separated by a blank line, so use the Find box to jump to a phrase from your page, then copy that section.
- Screenshot the page. Zoom in so the text is sharp, take a screenshot of that page and upload it to the Image to Text tool. This works for scans and text-based pages alike, as long as the image is 5 MB or smaller. Clear images are read in your browser.
What happens to formatting
Set your expectations here, because this surprises people. A PDF records where each piece of text sits on the page. It doesn't record that something is a paragraph, a heading or a table cell, so any tool has to work those out.
Our tool returns plain text. On a text-based PDF, the pieces of text on a page are joined with spaces and pages are separated by a blank line. Bold, columns and table grids don't carry over. Three buttons help tidy the result: Clean formatting, Remove line breaks and Convert to paragraph. The .docx download gives you editable text in Word, without the original layout.
If layout matters more than speed, try Google Docs first, keeping in mind the list of what Google says it won't detect.
Which method should you use?
- One passage from a text-based PDF: copy it from your browser. It's the quickest route, and nothing is uploaded.
- A whole text-based PDF: use the PDF to Text tool, so you can search, clean and download the result in one place.
- A small scan, if you already use Google Drive: Google Docs is convenient.
- A scan of up to about 3.7 MB, with no account: use our tool.
- One page of a large scan: screenshot it and use Image to Text.
Common problems and fixes
| Problem | Likely cause | What to do |
|---|---|---|
| Google Docs won't convert the file | The file is over 2 MB | Use a smaller file, or try our tool |
| Pasted text is gibberish | The PDF's fonts don't map to real characters | Screenshot the page and run it through OCR |
| Some pages are missing | A mixed PDF. If the file has real text anywhere, our tool treats it all as text-based and skips scanned pages | Screenshot the missing pages and use Image to Text |
| "Extraction failed" | Often a password-protected file, which isn't supported | Remove the password if you own the file, then retry |
| Words appear out of order | Columns or tables | Copy the sections separately, or check against the original |
| The tool stops accepting files | You've reached the limit of 10 extractions an hour | Wait for the counter to reset |
Whichever method you use, compare names, numbers and dates with the original before relying on them. OCR can misread characters, especially on faint or skewed scans.