Image to text (OCR)

OCR — optical character recognition — turns text inside an image (a screenshot, a photo of a document, a scanned page) into actual editable text. It's one of the most consistently useful image operations because text trapped in an image is text you can't search, copy, or edit.

Most OCR services require uploading your images to a server, which is a problem when the images contain sensitive content (which they usually do — contracts, receipts, business cards, ID cards, medical documents). Our OCR will run entirely in your browser using Tesseract.js with WebAssembly.

When you'll need this

Common use cases:

• Extract text from screenshots. Copy code from a screenshot of a tweet, copy data from a screenshot of a spreadsheet, copy quotes from a screenshot of a book.
• Digitize printed documents. Photograph an old printed page and get editable text.
• Read receipts and invoices. Extract line items, totals, dates for expense tracking.
• Process business cards. Quick way to get a contact into your address book.
• Translate text in images. Extract first, then translate the result.
• Accessibility. Convert images of text to actual text that screen readers can read.

We'll support around 100 languages on launch via Tesseract's language packs. Each language pack adds ~5 MB to the download but only the ones you use are fetched.

Private by design

Like everything else on pictoolkit, this tool will process your images entirely in your browser. No upload, no server-side processing, no temporary storage. Your files stay on your device.

While we finish this one, you can use any of our live tools — compress, convert, resize, HEIC to JPG, and our full format conversion suite are all available now.