Each page is rendered and passed through the tesseract.js OCR engine (English by default). You get the recognised text as a file, and an optional PDF with an invisible, selectable text layer positioned over the scan.
OCR a scanned PDF
Beta — see limitations belowExtract text from scanned pages with tesseract.js and export a text layer.
OCR a scanned PDF: Extract text from scanned pages with tesseract.js and export a text layer. Add your file to the tool above, apply it, and download the result — free, in any browser, with no sign-up.
- 100% free — no sign-up
- Files processed privately
- Fast — no upload wait
- No file-count limit
About this tool
How to use OCR PDF
1. Add your file
Drop your file onto the OCR PDF workspace above, or click to browse for it.2. Choose the options
Adjust the settings for ocr pdf — the defaults work for most files.3. Download the result
The result is generated in your browser — nothing is uploaded. Click Download to save it.
Frequently asked questions
How accurate is the OCR?
tesseract.js works well on clean, high-contrast scans of printed text. Handwriting, low-resolution scans, and unusual fonts reduce accuracy. Rendering at a higher scale helps.
Which languages are supported?
English is bundled. Other tesseract language packs can be enabled, but each adds a download the first time it runs.
Is OCR PDF free to use?
Yes. OCR PDF is completely free with no watermark, no account, and no file-count limit.
Are my files uploaded to a server?
No. Processing happens in your browser using WebAssembly and the Canvas API. Your documents never leave your device.
Which browsers are supported?
Any modern browser — Chrome, Edge, Firefox, or Safari. No extension or desktop app is required.
Related PDF tools
Looking for something else? See all PDF tools.