Search and copy from a scan
After OCR, the words on a scanned page are stored in a text layer. You can find a phrase with the search box in your PDF reader, select a line and copy it into another document.
OCR PDF
or drop a file anywhere here
Files other than PDF are converted first. Signing in raises the size limit.
Or start from a blank IRS form
Browse all IRS forms →- You add
- One scanned PDF
- You set
- The document languages
- You get
- The PDF with a searchable text layer
Running OCR on a PDF turns the picture of each scanned page into text you can search, select and copy. OCR stands for optical character recognition. This tool reads a PDF and returns a PDF in which image-only pages carry a text layer. You tick the languages the document is written in: English, German, French, Portuguese or Chinese (Simplified). Pages that already contain text are left as they are. The result is always a PDF. To get an editable document, run OCR here first and then convert the searchable PDF to Word.
What can you do when you run OCR on a PDF?
Tell it which languages to read
The Document languages list has five boxes: English, German, French, Portuguese and Chinese (Simplified). English is ticked when the options appear. Tick every language used in the document and leave the others clear, because more languages means slower OCR.
Keep pages that already have text
The job skips any page that already holds text and works on the image-only pages. A file that mixes typed pages with scanned inserts comes back with the typed pages as they were.
Prepare a scan for conversion
A converter can only export text that exists in the file. Give a scan its text layer here, then pass the searchable PDF to the PDF to Word page.
What OCR keeps and what it changes
Stays the same
- Every page, in the same order, with its scanned picture.
- Pages that already hold text. They are skipped and copied through as they are.
- The file on your device. You download a new PDF.
Changes
- Each page that has no text gets a hidden text layer, so you can search, select and copy its words.
- The words are read in the languages you ticked, and only those.
- A digital signature on the PDF stops being valid, because OCR cannot keep a signature intact while it adds text.
- The result is one PDF. The recognized text is not returned as a separate file.
Which OCR languages to tick
| Your scan has | Tick | Good to know |
|---|---|---|
| Text in English | English | It is ticked when the options open. |
| Text in one other language, such as German, French, Portuguese or Chinese (Simplified) | That language | Clear English if no page uses it. |
| Pages that mix two languages | Both languages | The choices are passed to the OCR engine together, so one page can be read in both. |
| A language that is not in the list | Nothing to tick | The list holds the five languages the OCR engine ships with. Use another OCR program for other languages. |
How to OCR a PDF in 3 steps
Add the scanned PDF
Drag the scan onto the upload box, or click the box and pick the file.
Tick the document languages
Under Document languages, tick each language used in the scan. English is already selected.
Run OCR and download
Press Run OCR PDF and let the progress bar reach the end. Download the file and try selecting a word to confirm the text layer is there.
Why run OCR on indox?
You see what the job will do before you start. There is one choice to make, the language list, and the rest is fixed: scanned pages get a text layer, pages with text stay as they were, and you receive a PDF. A single scan runs without an account. Your upload is sent over HTTPS when you press Run, and free-tier files are removed automatically within 24 hours after the job finishes. The page is also direct about what OCR cannot do here. It reads five languages, it will not redo an existing text layer, and it does not check its own accuracy. That lets you judge whether the tool fits your scan before you upload anything.
- Searchable PDF out, same pages in
- Five document languages to choose from
- Pages with text are left alone
- No sign-in for a single scan
- Scans on the free tier deleted within a day
What are the limits of OCR on a PDF?
- Size
- The free tier has a file size limit. Signing in raises it, and paid plans raise it further.
- Formats
- This tool works on PDF files. To change a file's type, use the PDF converter.
- Privacy
- Files are encrypted in transit and processed on indox infrastructure.
- Deletion
- Free-tier files are deleted automatically within 24 hours after the job finishes. Signed-in output stays in your history for your plan's retention period.
Limits worth knowing
- Five languages are offered: English, German, French, Portuguese and Chinese (Simplified). This page has no option for any other language or script.
- The tool does not check or score its own accuracy. Proofread names, numbers and dates before you rely on the text.
- A page that already has a text layer is skipped, even when that layer is wrong. This tool cannot repair poor OCR made by a scanner or another program.
- The form has no controls to straighten, rotate or clean a scan. Turn sideways pages upright before you upload.
- Remove the password from a protected scan before you upload it, because the form has no place to type one.
When the OCR result is not what you expected
| What you see | Why | What to do |
|---|---|---|
| Nothing is searchable after the run | The pages already had text and were skipped, or the OCR step failed and the job returned your PDF as it was | Search for a word from a scanned page. If nothing is found, run the job again. A page that already has a wrong text layer stays as it is. |
| Words come out garbled | The scan is low resolution, tilted or faint, or the page is in a language you did not tick | Rescan at a higher resolution, straighten the page and tick every language in the document. Tesseract works best on images of at least 300 dpi. |
| Sideways pages read as nonsense | The form has no rotate setting | Turn the pages upright with Rotate PDF, then run OCR again. |
| The signature shows as invalid | OCR rewrites the file, and a rewritten file cannot keep its digital signature | Keep the signed original as the record and use the OCR copy only for searching. |
| The job cannot open the file | The PDF asks for a password and the form has no place to type one | Remove the password with Unlock PDF, then run OCR on the open copy. |
What do the free and paid plans add?
Free
- All converters
- Daily credit refresh
- Email support
Basic
- Access to all features
- AI Extract
- Priority conversions
- Email support
Highest priority
Premium
- Everything in Basic
- Highest priority queue
- Larger file limits
- Priority support
Pay-As-You-Go
- No subscription
- Credits never expire
- Opt-in auto-refill
- Access to all features
What else can you do with a PDF?
Who runs OCR on PDFs?
Office staff with scanned paperwork
Signed contracts and letters come off the scanner as pictures. OCR lets you search them for a name, a date or a clause.
Students and researchers
A scanned chapter or journal article cannot be searched or quoted. With a text layer you can find a term and copy a passage into your notes.
Bookkeepers and accountants
Scanned invoices and receipts are hard to look up later. After OCR you can search each file for an invoice number or a supplier name.
When to run OCR
- You scanned a contract and need to find one clause.
- You want to copy a paragraph out of a scanned article.
- You plan to convert a scan to Word and need real text in it first.
- A German, French or Portuguese letter arrived as a picture-only PDF.
- Your PDF mixes typed pages and scanned inserts, and only the inserts need text.
Where can I read more?
Sources
- OCRController.javaStirling-PDF source code, v2.14.2
- Stirling-PDF base image Dockerfile, OCR languagesStirling-PDF source code, v2.14.2
- Advanced features, skipping pages that already have textOCRmyPDF documentation
- Cookbook, digitally signed PDFsOCRmyPDF documentation
- Improving the quality of the outputTesseract documentation
Facts checked on October 16, 2026
Do it all with indox
No tool matches that search.
OCR PDF FAQs
What does OCR do to a PDF?
It reads the letters on each scanned page and adds them as a text layer. The page keeps its picture, and you can now search, select and copy the words.
Which languages can the tool read?
English, German, French, Portuguese and Chinese (Simplified). Tick each one that appears in your document.
Should I tick all five languages to be safe?
No. The language field warns that more languages means slower OCR, so tick only the ones your document uses.
What happens if I clear every language box?
The job falls back to English.
My PDF already has selectable text. Will OCR run again?
No. Pages that already contain text are skipped, and only image-only pages are processed.
Can I get a Word or text file from a scan?
This tool returns a PDF. Run OCR first, then convert the result on the PDF to Word page.
How do I know the OCR worked?
Open the download and try to select a word or search for one. If nothing can be selected, the OCR step most likely did not run and the job returned your original, so run it again.
Do I need an account to run OCR?
No. A single job works without an account, and signing in or moving to a paid plan raises the limits.