How to use the Image to Text (OCR) tool
- Choose one or more images, or a scanned PDF.
- Select the language of the text. For mixed documents, pick a combination such as English + Russian.
- Click “Extract text”. The first run loads the OCR engine, which takes a few seconds.
- Copy the text or download it as a .txt file.
What is OCR?
OCR (optical character recognition) reads letters in an image and turns them into real text that you can copy, edit, search and translate. It saves retyping receipts, printed letters, book pages, slides photographed in a lecture or text inside a screenshot.
This tool uses Tesseract, a well-known open-source OCR engine originally developed at HP and later maintained with support from Google, running in your browser.
Tips for the best results
- Use sharp, well-lit images. Blurry phone photos are the most common cause of errors.
- Photograph the page straight on, not at an angle, and fill the frame with the text.
- Choose the correct language: it improves accuracy a lot, especially for accented letters and Cyrillic.
- Printed text works much better than handwriting, which OCR engines generally read poorly.
- For multi-column layouts, check the order of paragraphs in the result.
Privacy
Recognition runs on your own device with an OCR engine compiled to WebAssembly. Your documents are never uploaded. That makes it suitable for invoices, ID documents and contracts.
Frequently asked questions
Is the OCR free?
Yes, with no page limits or sign-up.
Which languages are supported?
English, Russian, German, French and Spanish, plus combinations of English with each of them. More languages will be added.
Can it read handwriting?
Only very neat block letters. The engine is designed for printed text.
Why is the first run slower?
The OCR engine and the language data (2 to 4 MB) are downloaded once. Later runs start immediately.