OCR

Image to Text (OCR)

Pull the text out of a screenshot, scan or photo. This OCR tool recognises text in 12+ languages and gives you editable, copyable text — your image is read on your device, and only the language model is downloaded.

Upload an image

or drag & drop — a screenshot, scan or photo of text

How to extract text from an image

Three simple steps — all on your device.

  1. 1

    Upload an image

    Drop in a screenshot, scan or photo that contains text.

  2. 2

    Pick the language & extract

    Choose the text's language and click Extract — the model downloads once, then reads your image locally.

  3. 3

    Copy or download

    Edit the recognised text if needed, then copy it or download it as a .txt file.

Why use Image to Text on PDF for Joy

12+ languages

English, Spanish, French, German, Hindi, Arabic, Chinese, Japanese and more.

On-device OCR

Your image is read locally in the browser — it is never uploaded to a server.

Editable output

Fix any misreads in the text box, then copy or download as .txt.

100% private, on-device

Your files are processed entirely in your browser. Nothing is uploaded to or stored on our servers, so sensitive documents stay yours.

Private OCR that reads your image on your device

Optical character recognition (OCR) turns pictures of text into real, editable text. Most online OCR tools upload your image to their servers. This one runs the recognition in your browser, so screenshots of private messages, scanned documents and photos of notes never leave your device.

Choose the language, and the recognition model downloads once from a CDN and is cached for next time — then the OCR itself runs locally on your image. Copy the result or download it as a text file.

Private by design. Every file stays on your device — nothing is uploaded to or stored on our servers.

Frequently asked questions

Is my image uploaded for OCR?

No. The recognition runs in your browser on your device. Only the language model is downloaded (once) from a CDN — your image itself is never uploaded.

Which languages are supported?

12+ including English, Spanish, French, German, Portuguese, Italian, Dutch, Hindi, Arabic, Russian, Chinese (Simplified) and Japanese.

Why does it take a moment the first time?

The language model (a few MB) downloads on first use and is cached afterwards, so subsequent scans in that language are faster.

Is it really free?

Yes. The core tools are completely free with no signup, no watermark and no daily limits.