OCR PDF

Make scanned PDFs searchable.

Import your scans

PDF
Merge PDF Combine PDF files and images in the order you need. Split PDF Split a PDF into separate files. Extract PDF Pages Save chosen pages as a new PDF. PDF to Images Save PDF pages as JPG, PNG or WebP. Images to PDF Turn images into one PDF. Convert to PDF Turn documents, spreadsheets, presentations and images into PDF.

How to make a scan searchable

Docs

Drop one or more scanned PDFs, or select them.

OCR

Searchable scans

The words go into the PDF as invisible text over each page, so you can search, select and copy them in any PDF viewer.

Every language Tesseract reads

Latin, Cyrillic, Arabic, Chinese, Japanese, Korean, Indic scripts and more. The script of the first page decides which languages are chosen for you.

Pages stay as they are

The scanned pictures and page count stay unchanged. Pages that already have text are skipped unless you enable re-recognition.

Show 2 moreShow less

Small first download

Nothing is downloaded until you use OCR. Then only the languages you need, with their size shown first. They are kept for next time.

On to Word

A searchable PDF can go straight on to PDF to Word or PDF to text without uploading it again.

Frequently asked questions

How do I make a scanned PDF searchable?

Drop your scan, check the languages, then select Recognize and Download.

Which languages can it read?

All 123 languages available in the Tesseract models. Open the language selector to add languages that were not suggested.

Does it change how my pages look?

No. The text is invisible and lies over the page picture, which stays exactly as it was.

Why is a file marked Has text?

Every page already contains machine-readable text. Enable re-recognition to process those pages again.

Show 2 moreShow less

Are my files uploaded?

No. The files stay in your browser and never leave your device.

Is it free?

Yes. No account and no usage limits.

Technical details

Engine

Tesseract 5 with its LSTM models (tessdata_fast) runs as WebAssembly through tesseract.js in a background worker in your browser.

Languages

A language pack is fetched the first time you use it, checked against the published file and kept on your device. Script detection uses Tesseract OSD on the first page that needs reading.

Show 2 moreShow less

Text layer

Each word is written as invisible text over its place on the page with our own glyphless font, so search, selection and copy follow the scan.

Limits

Vertical Chinese and Japanese columns, handwriting and pages turned on their side are read poorly. Pages are read at 300 dpi.