How to ocr text
- Add your file. Click the drop zone or drag a scanned PDF or image onto the page.
- Choose a language. Select the language the document is written in.
- Run OCR. Text recognition happens on your device using a downloaded model.
- Review & download. Extracted text appears ready to copy or save as a file.
Why do this locally instead of online?
Documents that need OCR are often scans of things people specifically don't want floating around unencrypted — old contracts, ID cards, handwritten notes. This tool runs the recognition engine itself on your device rather than sending the image to a server to be read.
Frequently asked questions
Is this tool really fully offline?
The recognition engine runs on your device, but the first use downloads a small language model, cached afterward. Your document is never uploaded.
How accurate is browser-based OCR?
For clear, typed text it performs well. Handwriting or poor lighting will reduce accuracy.
Does this produce a searchable PDF, or just plain text?
This tool extracts plain text for you to copy or download.
Can I OCR a photo instead of a scanned PDF?
Yes — JPG and PNG images are supported directly.