Image to Text (OCR)
Pull the text out of a screenshot, scan or photo and turn it into editable text. Runs entirely in your browser with on-device OCR — your image is never uploaded.
OCR runs in your browser — your image is never uploaded.
How it works
This tool runs Tesseract, the well-known open-source OCR engine, compiled to WebAssembly so it executes inside your browser tab. When you click Read text, the engine and the trained data for the language you picked are fetched once from a public CDN and cached. Your image is then handed to the engine locally — it is never sent over the network — and the recognised characters are written back into an editable text box. A progress bar reflects the engine’s own status events as it loads the model and scans the page. Because the heavy code is loaded only when you actually run OCR, the rest of the site stays fast.
How to extract text from an image
- Drop a JPG, PNG or WebP image, or click to choose one.
- Select the language of the text in the image.
- Click Read text and wait for the progress bar to finish.
- Review and correct the extracted text in the box.
- Copy it or download a
.txtfile.
Get the best results
OCR rewards a clean source. For the highest accuracy, use an image where the text is sharp, high-contrast, roughly horizontal and not too small — a screenshot, a flat scan or a steady photo of a printed page. Crooked, blurry, low-light or low-resolution images, decorative fonts and handwriting are much harder, and the engine may misread or skip characters. Picking the correct language matters too, because it loads the right character set and dictionary. Whatever you feed it, treat the output as a draft: read it through and fix any slips before you use it. If your source needs straightening or sharpening first, the rotate & flip and crop tools can help.
Why run OCR in your browser
The images people most want to digitise — an ID, an invoice, a contract, a private note — are exactly the ones you would not want to upload to an unknown server. Browser-based OCR removes that risk: the recognition happens on your own machine, so the picture never leaves it. The honest trade-off is the first-run download. The first time you use a language, the engine and its model (a few megabytes) are fetched and cached; after that the tool is fast and works offline, and the privacy guarantee — that your image is not uploaded — holds on every run, first or hundredth.
Related tools
Frequently asked questions
Is my image uploaded for OCR?
No. Recognition runs in your browser using a WebAssembly build of the Tesseract engine, so the picture itself is never sent to a server. The only thing downloaded is the engine and the language model, fetched once from a public CDN; your image stays on your device.
Why is the first run slow?
The first time you use a language, the browser downloads the OCR engine and that language’s trained data — a few megabytes in total. After that it is cached, so later runs start quickly and work offline. Switching to a new language downloads only that model.
Which languages are supported?
English, Spanish, French, German, Italian and Portuguese are offered directly, plus Dutch, Polish, Russian and Turkish. The underlying engine supports many more scripts; the relevant trained data is downloaded on demand when you choose a language.
How accurate is the text extraction?
Accuracy depends on the image. Clean, high-contrast, straight, reasonably large text — like a scanned document or a screenshot — comes out very well. Photos with skew, glare, low resolution, handwriting or unusual fonts are harder, so always proofread and edit the result before relying on it.
Can I use the extracted text commercially?
Yes. The tool is free to use and the OCR engine it uses is open source under the Apache-2.0 licence, which permits commercial use. The text it produces is yours to copy, edit and reuse.
Image Tools runs OCR entirely within your browser. Your image is never uploaded. The first run downloads the open-source engine and language model, which are then cached for offline use.