Skip to content
Image To Text

Free online OCR — no signup

Image to text in 30 languages

Image To Text is a free online OCR tool that turns a picture of text into text you can edit, search and paste. It takes no signup, reads JPG, PNG, WebP, GIF, BMP and AVIF in 30 languages, and handles up to 20 images in a single batch.

Add your images, pick the language, copy the text — three steps, a few seconds each, nothing to install.

  • 100% free
  • No signup
  • 30 languages
  • Up to 20 images
  • Works offline
Line breaks

How to extract text from an image

Three steps, a few seconds each. No account, no software, no upload wait.

  1. Add your images

    Drag a photo, scan or screenshot onto the drop zone, click it to browse your device, or press ⌘V / Ctrl+V to paste an image straight from the clipboard. Up to 20 images go in at once and are read one after another.

  2. Set the language

    Choose the language the text is written in — English is preselected, and 30 are available. Picking the right one matters most for accented Latin scripts and for Arabic, Hindi, Thai, Japanese, Korean and Chinese, which need their own character models.

  3. Copy or save the text

    Each result opens in an editable box with a confidence score, so you can fix a stray character before you take it. Hit Copy text to put it on the clipboard, Download .txt to save it, or grab every result in one file with Download all.

Technical specifications

Accepted image formatsJPG/JPEG, PNG, WebP, GIF, BMP, AVIF
OutputEditable plain text — copy to clipboard or download as .txt
Recognition languages30, from English and Spanish to Arabic, Hindi and Japanese
Batch limit20 images at a time, 25 MB per image
OCR engineTesseract 5 (LSTM) compiled to WebAssembly, about 3 MB
Language pack size0.5–4 MB per language, downloaded once and cached
Processing locationYour browser — no image is uploaded to a server
PriceFree, no signup, no watermark, no usage cap
Works offlineYes, once the engine and language pack have loaded

Frequently asked questions

How do I convert an image to text?
Drop the image onto this page, choose the language, and the recognised text appears under the thumbnail a few seconds later. There is nothing to install and no account step in between — the whole flow is add image, pick language, copy text.
Is this image to text converter free?
Yes, and there is no usage cap. No signup, no trial period, no watermark on the output, and no limit on how many batches of 20 images you run in a day.
Which image formats can it read?
JPG/JPEG, PNG, WebP, GIF, BMP and AVIF, at up to 25 MB per file. That covers phone photos, screenshots and scans saved out of almost any app. A PDF is not an image, so open the page and screenshot it instead.
How many languages does it recognise?
30 languages, including English, Spanish, Portuguese, French, German, Russian, Ukrainian, Turkish, Arabic, Hebrew, Hindi, Thai, Vietnamese, Japanese, Korean and both Simplified and Traditional Chinese. Each one has its own trained model that loads the first time you select it.
Can I paste a screenshot instead of choosing a file?
Yes — copy a screenshot and press ⌘V on a Mac or Ctrl+V on Windows and Linux anywhere on this page. Pasted images are renamed pasted-1, pasted-2 and so on, so a run of several screenshots stays easy to tell apart in the list.
How accurate is the extracted text?
Clean printed text at a reasonable resolution usually comes back at 90% confidence or better, and every result carries its own score. Accuracy drops on blurry photos, tight line spacing, decorative fonts and text laid over a busy background — small images are enlarged automatically before recognition to claw some of that back.
Can it read handwriting?
Not reliably. The model is trained on printed and typed characters, so neat block capitals sometimes come through while ordinary cursive rarely does. For handwritten notes, treat whatever it returns as a rough draft.
Are my images uploaded to a server?
No. The recognition engine is downloaded to your browser once and then runs there, so the image data stays on your device and no copy of it is sent anywhere, stored or logged.

About this tool

Image To Text exists for the small, annoying moments: a phone photo of a whiteboard, a screenshot of an error message, a scanned page you need three sentences out of. Retyping any of them takes longer than the recognition does.

The engine is Tesseract 5, the long-running open-source OCR project, compiled to WebAssembly so it runs at close to native speed inside a browser tab. Each image is decoded, scaled into the resolution band the model was trained for, and passed through the language model you selected. What comes back is plain text plus a confidence score — which is why the result lands in an editable box rather than a locked-off panel.

The first scan is the slow one: the engine weighs about 3 MB and each language pack another 0.5–4 MB. Both are cached afterwards, so the second image and every one after it starts recognising straight away.

Looking to extract text from an image? Same engine, same privacy — just a different starting point.

What happens to your images

Nothing leaves the tab. Decoding, recognition and the text output all run on your own machine, so no image is transmitted, stored or logged, and closing the page releases the in-memory copies. The only things this site downloads are the engine and the language pack you picked.