Skip to content
PDFLovePDFby CodingMarble

OCR a PDF and make it searchable

Turn a scanned or photographed PDF into a searchable PDF. The pages look the same, but you can now find, select and copy the text. Free, with no sign-up.

or drop a file here · you can also paste

Sent over an encrypted connection and deleted right after processing.

How to OCR a PDF for free

  1. 1

    Select the scanned PDF

    Tap Select PDF file or drop the document on the page. One file at a time, up to 100 MB.

  2. 2

    Choose the language

    Under Language in the document pick English, Hindi, English + Hindi, Spanish, French or German. Choose the language printed on the pages, not the one you speak.

  3. 3

    Make it searchable

    Click Make searchable. The file is uploaded over a secure connection and each scanned page is read. Long documents can take a minute or more.

  4. 4

    Download the result

    Download the searchable PDF. Open it and press Ctrl+F to check that words can be found.

What OCR does to a scanned PDF

You scan a ten-page agreement, open it, press Ctrl+F and type a name. Nothing is found, even though the name is right there on the screen. You try to copy a paragraph and the whole page gets selected like a photo. That is because it is a photo. A scanner or phone camera saves pictures of pages, and a picture of a word is not a word.

OCR stands for optical character recognition. Software looks at the picture, works out which letters the shapes are, and writes those letters into the file. This online OCR tool does that for every scanned page and hands back a searchable PDF.

Searchable PDF meaning, in plain words

A searchable PDF is a scan with a hidden layer of real text placed exactly on top of the picture. You still see the original image, with its stamps, signatures and coffee stains. But when you drag the mouse over a line, you are selecting the invisible text that sits over it. That is the whole OCR PDF meaning: picture on the bottom, recognised text on top.

You may see the same job called a searchable PDF maker, an OCR PDF generator or a searchable PDF converter. They all mean the same thing.

Searchable PDF vs a normal scanned PDF

Is a searchable PDF a different format? No. There is no special searchable PDF format or OCR PDF format. Both files are ordinary PDFs with the same .pdf ending, and both open in every reader. What makes a PDF searchable is only the text layer inside it. A PDF exported from Word already has one. A PDF that came from a scanner does not, until OCR adds it.

  • Search: Ctrl+F finds words in a searchable PDF, and finds nothing in a plain scan.
  • Copy: you can select and copy a sentence, an account number or an address.
  • Find the file later: Windows, macOS, Google Drive and email search can look inside it.
  • Accessibility: screen readers can read the pages aloud.
  • Looks: no change at all. The page image is kept as it is.

OCR PDF full form, an example and a word about AI

The OCR PDF full form is optical character recognition applied to a PDF. Optical, because it works from the picture of the page. Character, because it deals with letters and digits one at a time. Recognition, because it decides which letter each shape is. Put together, it is the step that turns a photograph of a page into text a computer can use.

An OCR PDF example, before and after

Take a scanned electricity bill. Before OCR, Ctrl+F for the consumer number finds nothing, and dragging the mouse selects the whole page as one picture. After OCR, the same file looks identical, but Ctrl+F jumps to the consumer number, the amount can be copied into a spreadsheet, and Google Drive finds the bill when you search for the month. A three-page scan becomes searchable in well under a minute, and that is a typical OCR PDF example: same look, new abilities.

Is this OCR PDF AI?

The engine uses a trained neural network to recognise characters, which is a form of machine learning, so in that sense the OCR is AI. What it is not is a chatbot: it does not summarise, rewrite or send your text to a third-party AI company. It runs on our own server, reads the shapes, writes the text layer and forgets the file. If you want the text in another language afterwards, the Translate PDF tool runs on the same server with the same rule.

A searchable PDF converter that leaves good pages alone

Many real documents are mixed. A report typed on a computer has three scanned annexures at the back. A loan file has digital bank statements and photographed receipts. This searchable PDF converter checks every page first. Only pages without a text layer go through OCR. Pages that already have real text are kept exactly as they are, so their sharp text is never replaced with a rougher guess.

If every page in your file already has text, there is nothing to do, and you get the same file back unchanged. So it is safe to run a document through the tool when you are not sure whether it is a scan.

Six languages, including Hindi

The OCR engine needs to know which alphabet to expect. You can choose English, Hindi, Spanish, French or German. There is also an English + Hindi option for pages that mix both, such as government circulars, school certificates and many bank forms in India. Other scripts, for example Tamil, Bengali, Arabic or Chinese, are not available yet, and picking the wrong language gives poor results.

OCR PDF guide: how to get high quality results

OCR can only read what the scan shows clearly. Clean printed text scanned straight is recognised very well. A dim, tilted phone photo is not. This short OCR PDF guide covers what matters most.

  • Scan at 300 dpi if you use a flatbed scanner. High quality OCR needs detail. PDF pages below 150 dpi are usually too coarse, and small print breaks first.
  • Keep pages straight and the right way up. If the file is sideways, rotate the PDF and save it before OCR.
  • Use even light with no shadow across the text when you scan paper to PDF with your phone. The Clean grayscale look suits OCR well.
  • Run OCR before heavy compression. If you need a small file, compress the PDF afterwards. The text layer survives.
  • Do not expect much from handwriting. Neat block capitals are sometimes read. Normal handwriting mostly turns into nonsense.

Check the result before you rely on it

OCR makes small mistakes even on good scans: 0 for O, 1 for l, rn for m. For reading and searching that hardly matters. For figures, such as account numbers, amounts and dates, compare what you copied with the page image before you use it.

OCR PDF to Word, Excel or plain text

A searchable PDF is often only the halfway point. Once the text exists, other tools can use it.

From OCR PDF to Word

A scanned file sent straight to a Word converter gives you a document full of pictures that you cannot type into. Do it in two steps. Run OCR here first, then convert the PDF to Word. You get a .docx with editable paragraphs and tables. Used together, the two online tools work as a free OCR PDF to Word converter. Expect to tidy line breaks and spacing, as with any scan.

From OCR PDF to Excel

Printed statements, old invoices and mark sheets follow the same route. After OCR, convert the PDF to Excel and the tables are split into rows and columns. Pairing the tools like this gives you an OCR PDF to Excel converter, and it works just as well for a searchable PDF to Excel that someone else already ran OCR on. Check the digits carefully, because one misread number spoils a total.

How to convert a scanned PDF to text

This tool does not produce a .txt file. Getting from a scanned PDF to text is still simple. Open the searchable PDF, press Ctrl+A to select everything and Ctrl+C to copy, then paste into Notepad, an email or a chat. For a single paragraph, select just those lines. That is how to convert a scanned PDF to text for free without a separate scanned PDF to text converter.

Is there a free OCR PDF editor online?

OCR does not make the words on the page editable in place. The visible page is still an image. So an OCR PDF editor online, free or paid, is really two tools: OCR to get the text, then something to change it. To change the wording, go through Word as described above. To add things on top, such as a date, a tick mark or a signature, open the file in the Edit PDF tool, which is also free and runs in your browser.

How long does it take to OCR a PDF, and is there a limit?

Speed depends on the number of scanned pages and how much text they hold. As a rough guide, a page of normal printed text takes two to five seconds on the server. A 10-page contract is done in well under a minute; a 150-page scanned book can take five to ten minutes, and the page keeps working in the background while you wait. Pages that already contain text are skipped, so a mixed file is faster than its page count suggests.

Unlimited OCR: no page cap, no daily count

There is no cap on pages and no counter of files per day: OCR for PDF files here is unlimited, so run one file after another for as long as you need. The practical limits are one file at a time and 100 MB per file. If a scan is bigger than that, split the PDF with the Split PDF tool into parts first, OCR each part, then merge the searchable parts with Merge PDF back together.

Speed tips for large scans

Pick the single language that matches the pages, because English + Hindi runs two recognisers and takes longer. Do not compress a scan into a blur before OCR; reading a 300 dpi page is quicker than guessing a 100 dpi one. And keep the browser tab open until the download appears.

Where your file goes, and working offline

You can OCR a PDF in the browser you already have. It works the same in Chrome, Safari and Firefox, and with Edge on a Windows PC. The recognition itself is heavy work, too heavy for most phones, so it runs on our server and not on your device. The file travels over an encrypted HTTPS connection, is processed automatically with no person looking at it, and is deleted right after the result is delivered to you.

Can I OCR a PDF offline?

Not with this tool. It is an online OCR service, so it needs a connection to upload the PDF. There is no free download or desktop version either. If a document must never leave your computer, you need installed OCR software to OCR the PDF offline. For everyday papers, the delete-after-processing rule above is the protection we offer.

The other direction: searchable PDF to non searchable PDF

Sometimes you want the opposite. You are sharing a document and do not want the text copied or picked up by search. To turn a searchable PDF into a non searchable PDF, save the pages as JPG images at 144 or 216 dpi, then put the JPG files back into one PDF. The new file is pictures only. Keep in mind that anyone can run OCR on it again. To hide a detail for good, redact the PDF.

OCR PDF: questions and answers

How do I OCR a PDF?

Select the scanned PDF, choose the language printed on its pages, and click Make searchable. The tool reads each scanned page and adds an invisible text layer over the image. Download the new file and press Ctrl+F to test it. The pages look the same as before, but words can now be found, selected and copied.

Is this OCR PDF converter free?

Yes. It is free to use, with no account, no watermark on the result and no daily limit. You can make a searchable PDF online free as often as you need. The only limits are one file at a time and a maximum of 100 MB per file.

How can I tell whether my PDF is scanned or already searchable?

Open it and try to select one word with the mouse or your finger. If single words highlight, the PDF has a text layer. If the whole page is selected like a picture, or nothing is, it is a scan and needs OCR. Searching for a word you can see on the page with Ctrl+F is another quick test.

Will OCR change how my document looks?

No. The scanned page image stays where it is, including stamps, signatures and handwriting. The recognised text is added as an invisible layer on top. Pages that already contain real text are not touched, and a file in which every page already has text is returned unchanged.

Which languages can it read?

English, Hindi, English and Hindi together, Spanish, French and German. Pick the language that is printed in the document. For a page that mixes Hindi and English, use the English + Hindi option. Other languages and scripts are not supported yet, and choosing a wrong language produces wrong text.

Can it read handwriting?

Poorly. OCR is built for printed or typed characters. Clear block capitals on a form are sometimes recognised, but normal handwriting usually comes out as nonsense. The page still looks right because the image is kept. Only the hidden text will be unreliable, so searching handwritten notes will not work well.

Why are some words wrong when I copy the text?

The scan was probably unclear at that spot. Low resolution, tilted pages, shadows, tiny print, coloured backgrounds and stamps over text all cause mistakes. Scan again at 300 dpi with the page straight and evenly lit, then run OCR again. Always check numbers such as amounts and account numbers against the page.

My PDF is password protected. Can I still run OCR?

Remove the password first. If you know it, the Unlock PDF tool takes it off, and then the file can be processed here. Restrictions that only block copying or printing can be removed there too. You can add a password again afterwards with Protect PDF.

Can I search a scanned PDF after OCR on my phone?

Yes. The text layer is part of the file, so it works in any PDF reader on Android or iPhone. Use the search or magnifier icon in the reader app. You can run the OCR from your phone as well: pick the scan from your files, and download the searchable copy.

Why isn't my PDF searchable?

Because it has no text layer. A PDF made by a scanner, a photocopier or a phone camera stores each page as a picture, and a picture of a word is not a word, so search finds nothing. Sometimes a PDF exported from a design program is the same, if the text was outlined. Run the file through this OCR tool and the missing text layer is added.

How long does it take to OCR a PDF?

A few seconds per scanned page. A 10-page document is usually done in under a minute, and a 100-page one in a few minutes, depending on how dense the pages are and how busy the server is. Pages that already have text are skipped. Keep the browser tab open until the download button appears.

Is there a limit on pages or files?

No page cap and no daily count. The only limits are one file per run and 100 MB per file. For a bigger scan, split it into parts with the Split PDF tool, run OCR on each, then merge the searchable parts again with Merge PDF. Nothing is watermarked and no account is needed.

More PDF tools you might need