Image to Text

Retyping the text of a screenshot, a photographed letter or a picture of a slide is slow and easy to get wrong. Add the image here and optical character recognition (OCR) reads the printed words for you, ready to copy or to save as a text file. The recognition runs in your browser, so the picture stays on your device.

or drop files here

Processed in your browser – nothing is uploaded

How to use Image to Text

  1. Click “Select images” or drop one or more pictures onto the box.
  2. Choose the language of the text in the picture.
  3. Press “Recognise text” and wait while each image is read.
  4. Copy the text, or download it as a .txt file – one per image when you added several.

About this tool

The picture is decoded by your browser, placed on a white background and passed to Tesseract, an open-source recognition engine that runs here as WebAssembly. Small pictures such as screenshots are enlarged first, because the engine reads letters best when they are a few dozen pixels tall; very large photos are reduced to a size your device can hold in memory. The engine finds blocks, lines and words and returns them in reading order, keeping line breaks and the wide gaps between columns.

It works well on screenshots of documents, web pages and chats, scanned letters and invoices, book pages photographed straight-on in good light, and printed labels or signs with dark text on a light background. You can add several images at once: each gets its own .txt file, and all the text is shown together for copying.

It does not work on handwriting, and it struggles with text that is blurred, tilted by more than a few degrees or wrapped around a curved surface, with tiny text in low-resolution photos, stylised logos and pale text on busy photographic backgrounds. Tables are read row by row and come out as plain lines rather than a spreadsheet, and multi-column layouts can be mixed up. Text has to be upright, so turn a sideways picture with Rotate Image first. Always check numbers such as amounts, dates and account details against the picture before you rely on them.

A screenshot usually takes one to three seconds on a laptop and several times longer on a phone. The first run downloads the engine (about 1.5 MB) and the language you chose (0.7–3 MB) from this site; later runs use the copies in your browser’s cache.

Frequently asked questions

Which image formats are supported?

JPG, PNG, WebP, BMP, GIF (the first frame), AVIF, and HEIC or HEIF photos from iPhones. For a scanned PDF use OCR PDF instead, which keeps the pages and makes them searchable.

Which languages can it read?

English, Spanish, French, German, Portuguese, Italian, Dutch, Russian, Arabic, Urdu, Hindi, Simplified Chinese and Japanese. Choose the language of the text before you start. One language is used per run, so a page that mixes two scripts is only read well in one of them. Latin and Cyrillic print is read most reliably; Arabic and Urdu are the weakest, especially in calligraphic Nastaliq typefaces.

Can it read handwriting?

No. The engine is trained on printed type. Neat block capitals are sometimes recognised, but joined-up handwriting comes out as nonsense.

How do I get better results?

Use the sharpest and largest version of the picture you have, crop it to the text, make sure the text is upright and pick the right language. For photos of paper, flat and even light without shadows helps more than anything else.

Is the formatting kept?

Line breaks and wide gaps between columns are kept. Fonts, sizes, colours, bold type and table borders are not, because the result is plain text.

Are my images uploaded to a server?

No. The whole process runs inside your browser with JavaScript. Your images never leave your device, which also means the tool keeps working if you go offline after the page has loaded.

Is this tool really free?

Yes. There is no sign-up, no daily limit, no watermark and no paid tier. The site is supported by advertising shown on the page, not by charging for the tools.