Image to Text (OCR)

Turn photos, screenshots and scans into editable text in over 100 languages, right in your browser.

NewFreeNo uploadNo sign-up

Settings

Pick every language that appears in the image. Language data (2–15 MB) downloads once, then works offline.
Add an image to start.
Extracted text

How to use Image to Text (OCR)

  1. 1

    Add your image

    Drop a photo, screenshot or scanned page, choose it from your device or paste it with Ctrl+V.

  2. 2

    Pick the language

    Search the list and tick every language in the image. You can mix up to five, such as English and Japanese.

  3. 3

    Extract and copy

    Press Extract text, then edit, copy or download the result as a .txt file.

Free OCR that works in your language

Optical character recognition (OCR) reads the letters in a picture and turns them into real, editable text. This tool is built on Tesseract, the open-source OCR engine originally developed at HP and later maintained by Google, and it supports more than 100 languages and scripts: Latin alphabets such as English, Spanish, French, German, Portuguese, Romanian and Polish, as well as Cyrillic, Greek, Arabic, Hebrew, Devanagari, Chinese, Japanese, Korean, Thai and many more. Vertical Chinese, Japanese and Korean text is supported too.

Private by design

Most online OCR services upload your file to a server. Here the recognition engine runs inside your browser using WebAssembly, so the image never leaves your device. The only thing downloaded is the language data the first time you use a language. After that it is cached and the tool keeps working even offline, which makes it safe for receipts, contracts, ID documents and other personal files.

Tips for the best results

  • Use a sharp, well-lit image where the text is straight and not blurred.
  • Select only the languages that actually appear in the image. Extra languages slow things down and can reduce accuracy.
  • Keep Enhance image switched on for phone photos and small screenshots. It enlarges the image and boosts contrast before reading.
  • For signs, memes or posters with scattered words, choose the Sparse text layout. For a single line, such as a label or serial number, choose Single line.

What you can use it for

Copy a quote from a screenshot, digitise printed notes, pull an address from a photo, translate a menu by pasting the text into a translator, or make an old scanned letter searchable. Handwriting is recognised only when it is very neat, so printed text gives by far the best results.

Frequently asked questions

Which languages are supported?
More than 100 languages and scripts, including English, Spanish, French, German, Italian, Portuguese, Romanian, Russian, Ukrainian, Polish, Turkish, Arabic, Persian, Hebrew, Hindi, Bengali, Tamil, Chinese (Simplified and Traditional), Japanese, Korean, Thai and Vietnamese. You can combine up to five languages in one image.
Is my image uploaded anywhere?
No. Text recognition runs entirely in your browser. Only the language data is downloaded from a public CDN the first time you use a language, and your image is never sent to any server.
Why is the first run slower?
The first time you choose a language, its data file (usually 2 to 15 MB) is downloaded and cached. Later runs start almost instantly and work offline.
Can it read handwriting?
It is designed for printed text. Very neat handwriting may be recognised, but results with cursive or messy handwriting will be poor.
Is there a limit on how many images I can convert?
No. The tool is completely free with no limits, watermarks or sign-up.

Related tools

Image to PDFScreenshot to PDFCrop ImagePDF to Image