Image to Text — Extract Text from Images (OCR)

Drop a screenshot, a photo of a document or a scanned page and get the text out — editable, copyable and searchable. Recognition runs on your device, so private documents stay private.

Runs in your browserUpdated
Drop images or screenshots here, click to choose, or paste with Ctrl+V · PNG, JPG, WebP, BMP, GIF
Output

The OCR engine (about 7 MB, plus 3 MB per language) downloads the first time you add an image.

Your data never leaves your computer. All processing happens 100% locally in your browser.

How to use the Image to Text (OCR)

  1. Drop one or more images — or paste a screenshot with Ctrl+V.
  2. Pick the language of the text for the best accuracy.
  3. Wait a few seconds per image, then copy or download the text and fix any misread words.

How OCR works here

Optical character recognition finds lines of text in an image and recognises the characters with a neural network. This tool runs Tesseract, the open-source OCR engine originally developed at HP and later by Google, compiled to WebAssembly so it works entirely in your browser. Nothing is uploaded — useful for IDs, contracts, invoices and screenshots of private messages.

Getting the best results

  • Use a sharp, straight image — text at least 20 pixels tall, good contrast, no blur. Screenshots work best of all.
  • Pick the right language — accents and special letters are only recognised with the matching language data.
  • Crop to the text — photos with lots of background, tables or columns can mix up the reading order.
  • Handwriting is not supported well; OCR is built for printed text.

Multiple images and PDFs

Drop several images and their text is extracted one after another into the same result. For scanned PDFs, use PDF to Word with OCR turned on — it reads every page and builds a Word document.

Languages

English data is included. Other languages download their recognition data (2–4 MB) from the jsDelivr CDN the first time you choose them; the images themselves still never leave your device.

Tools that pair well with the Image to Text (OCR)

  • QR Code Scanner — Point your camera at a QR code, or drop an image or screenshot that contains one.
  • Image Compressor — Shrink photos and graphics by 50–90% without visible quality loss.
  • Word Counter — Paste or type your text to count words and characters live — plus reading and speaking time, keyword density and length checks for SEO titles, meta descriptions and social posts.
  • Barcode Generator — Make scannable barcodes for products, inventory, shipping and libraries — check digits are calculated for you, and you can download a sharp SVG for print or a PNG for documents.

Frequently asked questions

Is the image uploaded to a server?
No. The OCR engine runs in your browser; your images and the extracted text stay on your device. Only the engine and language data are downloaded.
How accurate is the text recognition?
For clear screenshots and scans of printed text, usually over 95% of characters are correct. Photos taken at an angle, small text, decorative fonts and handwriting give worse results.
Can it read text in other languages?
Yes — more than 20, including Spanish, French, German, Arabic, Hindi, Sinhala, Tamil, Chinese, Japanese and Korean. Choose the language before adding the image.
Can I extract text from a PDF?
If the PDF has selectable text, PDF to Word extracts it directly. For scanned PDFs, turn on OCR in PDF to Word, or convert pages to images with PDF to JPG and drop them here.
Can I install Image to Text (OCR) as a web app and use it offline?
Yes. Image to Text (OCR) is an installable web app (PWA) — installed from your browser, not downloaded from an app store. Click Install app at the top of this page in Chrome or Edge (Windows, macOS, Linux, ChromeOS) or on Android, use Share → Add to Home Screen on iPhone and iPad, or File → Add to Dock in Safari 17+ on a Mac. It then opens in its own window with its own icon, and because it runs entirely in your browser it keeps working offline. Firefox on desktop can’t install web apps. How installing works.

Related tools