Are my images uploaded anywhere?

No. Every tool runs inside your browser using Web Workers and WebAssembly. Files never leave your device, and there is no server, account or tracking.

Printed text in a sharp, straight image is read very reliably. Handwriting, tiny text, heavy skew and blurry photos lower the accuracy; a confidence score is shown for each image so you know when to check the result.

Which languages are supported?

English for now. Numbers and punctuation are read in any document.

Why is there a download the first time?

The recogniser (Tesseract, compiled to WebAssembly) and its language file are about 9 MB. They are fetched once from this site and kept in your browser, so later images are read without any download.

Can I keep the layout?

Choose “As in the image” to keep one line of text per line in the picture, or “Flow into paragraphs” to join the lines of each paragraph, which pastes cleanly into a document.

Image to text

Read the text in screenshots, scans and photos and copy it as plain text. The recogniser runs on your device, so the images stay private.

or drop files here, or paste from your clipboard

JPG, PNG, WebP, AVIF, GIF, BMP, SVG · processed on your device

How to image to text online without uploading

Turn a picture of text into text you can copy: screenshots, scanned pages, photos of documents, receipts and signs. The recogniser runs on your device, so the images never leave it. Drop several images at once and copy them together or one at a time, as separate lines or flowed into paragraphs.

1Drop screenshots, scans or photos, or paste one from the clipboard
2Wait a moment while the text is read on your device
3Copy the text, or save all of it as a .txt file

Frequently asked questions

Are my images uploaded anywhere?: No. Every tool runs inside your browser using Web Workers and WebAssembly. Files never leave your device, and there is no server, account or tracking.
How accurate is it?: Printed text in a sharp, straight image is read very reliably. Handwriting, tiny text, heavy skew and blurry photos lower the accuracy; a confidence score is shown for each image so you know when to check the result.
Which languages are supported?: English for now. Numbers and punctuation are read in any document.
Why is there a download the first time?: The recogniser (Tesseract, compiled to WebAssembly) and its language file are about 9 MB. They are fetched once from this site and kept in your browser, so later images are read without any download.
Can I keep the layout?: Choose “As in the image” to keep one line of text per line in the picture, or “Flow into paragraphs” to join the lines of each paragraph, which pastes cleanly into a document.