Image to text

Extract text from an image or screenshot with OCR, then copy it or save it as a .txt file

Need the image itself in another language? Our image translation service translates its text while preserving the layout.

Need text translation in images?

The image translator gives you the same image back with its text in another language and its original layout.

Try translation

How image to text works

Image to text (OCR) turns the words in a photo, scan or screenshot into text you can copy and edit.

  • The text is recognized in your browser by Tesseract.js, an open-source OCR engine, in the languages you select: up to 3 of 13.
  • The image never leaves your device: it is not uploaded to our server.
  • The first time you use a language, your browser downloads the OCR engine and that language's data from the jsDelivr CDN.

Free to use, with no account or sign-up. No watermark or logo is added to your result.

On a computer or a phone

The tools open in your browser: nothing to install and no account to create.

Computer

Recent versions of Chrome, Firefox, Safari and Edge.

Mobile

Safari on iPhone and iPad, Chrome on Android. Upload a photo from your gallery or your files.

Frequently asked questions

Everything you need to know about extracting text from an image.

How do I extract text from an image?

Upload a JPEG, PNG, WebP, AVIF or GIF image of up to 10 MB, or paste a screenshot with Ctrl+V. Choose the language of the text and click Extract text. The text appears in a box where you can correct it. Only the first frame of an animated GIF is read.

How do I copy text from an image?

Extract the text, then click Copy text to put all of it on your clipboard, or select only the part you need in the result box. You can correct misread words in the box first. Download .txt saves the same text as a plain text file.

Which languages can it read?

English, French, Spanish, German, Italian, Portuguese, Japanese, Korean, Simplified Chinese, Traditional Chinese, Arabic, Russian and Hindi. When an image mixes languages, select up to three of them, for example English and Japanese. Each extra language makes recognition slower.

Does it work with handwriting?

It is made for printed text, such as screenshots, scanned pages, signs and labels, and works best on sharp images with good contrast. Handwriting is often misread or skipped, so check the result carefully.

Is my image uploaded to a server?

No. The text is recognized in your browser with Tesseract.js, an open-source OCR engine, and the image is never sent to our server. The first time you use a language, your browser downloads its data from a public CDN (jsDelivr), along with the OCR engine, and keeps the language data for next time.

How do I translate the extracted text or the image?

This tool extracts text but does not translate it: paste the copied text into the translation tool you prefer. To get the image itself in another language, with the translated text placed back in the original layout, use our image translator.

Translator, API and guides

To translate one image, use the image translator; to translate images from your own code, use the API. The blog has guides, and our comparison with Google Translate explains when each tool fits.

Get your image back in another language

Upload a product photo, ad, menu or screenshot and get it back with its text in another language. Check names, numbers and prices before you share or publish it.

Text rewritten by an AI image model
Follows the original layout, fonts and colors
45 target languages
No credit card required
Free plan: 3 images a month