Image to Text (OCR)

Pull text out of a photo or screenshot — 100% free, runs in your browser

Click to choose an image

Screenshot, photo of a document, or a scan

Language of the text

Runs entirely in your browser, so the image never leaves your device. The first run downloads the language data, which takes a moment. Clear, straight, well-lit text gives much better results than a tilted phone photo.

About the Image to Text (OCR)

Optical character recognition turns a picture of text into text you can edit. It is useful for copying from a screenshot, digitising a printed page, or pulling numbers out of a document photo. Bengali, Hindi and English are all supported, and the recognition runs on your own device so the image is never uploaded.

How to use it

Choose a photo, screenshot or scan.
Select the language of the text — accuracy drops noticeably if this is wrong.
Press Extract Text and wait while the engine loads.
Copy the result and correct anything it misread.

Common questions

Why is the first run slow?

The recognition engine and language data download once, which takes a few seconds. Later runs are much faster.

How accurate is Bengali recognition?

Good for clean printed text, weaker for handwriting or complex conjunct letters. Always proofread the output.

How do I get better results?

Photograph straight on rather than at an angle, in even light, with the text filling the frame. A slightly blurry photo produces far more errors than most people expect.

Is my document private?

Yes. Everything happens inside your browser. Nothing is sent to any server.

Related tools

Want AI to do this 10x better? ⚡

AI Logo Designer • Marketing Copy • YouTube SEO — 100 free credits on signup

Try AI Studio Free →