What is Image/PDF → Text (OCR)?
Turn a photo of a page, a screenshot or a scanned PDF into editable text. This OCR tool recognises characters using a model that runs on your own device, so even confidential scans stay private — something no cloud OCR service can offer.
How to use Image/PDF → Text (OCR)
- Choose a scanned image or PDF containing the text you want.
- Pick the language of the document for the most accurate results.
- Click “Extract text”. The first run downloads a small language model, so allow a moment; later runs are much faster.
- Copy the recognised text or download it as a .txt file.
Why use this tool?
Private by design
Recognition happens on your device, so receipts, ID documents and contracts are never sent to a third-party OCR service.
Multiple languages
English, Hindi, Spanish, French, German, Arabic and Chinese are supported, each with its own trained model.
Works on scanned PDFs
Scanned PDFs have no text layer, so ordinary extraction finds nothing. OCR reads the characters out of the page image instead.
Frequently asked questions
Why is the first run slow?
The language model has to download once before recognition can start. Your browser caches it afterwards, so subsequent runs begin almost immediately.
How accurate is the OCR?
Clear, high-contrast, straight scans are recognised very accurately. Blurred photos, handwriting, unusual fonts and skewed angles reduce accuracy — retaking the photo in good light usually helps more than anything else.
Is my image uploaded for recognition?
No. The image stays on your device — only the recognition model is downloaded to you, never the other way round.