Image to Text (OCR)
Copy the text out of a screenshot, a scanned page or a photo of a document. Choose the language, check the result, and copy it or download it as a text file.
- Images stay on your device
- Free, no account, no watermark
- 10 languages
- Images stay on your device
- Copy or download TXT
Recognise
The first run downloads the OCR engine and language data (a few MB) from the jsDelivr CDN. Your images are not uploaded.
Images
Your images are read in your browser and are not uploaded. The free OCR engine and language files are downloaded from a public CDN the first time you use it. Privacy policy
How to use the Image to Text (OCR)
- Drop one or more images on the box, choose them, or paste a screenshot.
- Choose the language of the text.
- Click Extract text and wait for the progress bar. The first run downloads the language data.
- Check and edit the text, then copy it or download it as a .txt file.
What is OCR?
Your images are read in your browser and are not uploaded. The free OCR engine and language files are downloaded from a public CDN the first time you use it.
Free to use, with no sign-up, no limits on the number of files and no watermark added.
Built and maintained by SBGeeks, a digital marketing agency. Questions or problems? Contact us.
How it works
The OCR engine (Tesseract, compiled to WebAssembly) runs inside your browser. The first time you use a language, the engine and that language’s data, a few megabytes, are downloaded from the jsDelivr public CDN and cached by your browser. Your images are processed on your device and are not sent anywhere.
Small images are enlarged and converted to grayscale before recognition, which improves accuracy on screenshots and low-resolution photos.
Useful for
- Copying text from screenshots and images in emails
- Digitising printed documents and receipts
- Extracting quotes from photographed pages
- Getting editable text from a scanned PDF page saved as an image
Limitations
- Handwriting, decorative fonts and text on busy backgrounds are often misread.
- Accuracy depends on resolution: text should be at least 20 pixels tall.
- Columns and tables are read as plain lines; layout is not preserved.
- The first run needs an internet connection to download the engine and language data.
Questions
Are my images uploaded?
How can I get better results?
Can it read PDFs?
Related free tools
Our content team writes and structures pages that rank in search and answer real customer questions.