Image to Text

Text from a picture, without retyping it

A photo of a notice, a screenshot of an error message, a scanned letter, a slide from a talk: the words are right there, but you can’t select them. Retyping is slow and easy to get wrong. This tool reads the printed text in the picture with OCR (optical character recognition) and gives you a plain text file to copy, search or edit.

How to use it

1. Choose up to 10 images: JPG, PNG, WebP, GIF or BMP, each up to 25 MB.
2. Press Read the text. The first time, the page loads its reading engine; after that it starts at once.
3. Download each .txt file, or all of them as a ZIP.

Getting the best result

• Fill the frame with the text. Crop away the rest first if you can (Crop Image).
• Keep it straight and sharp. Hold the phone parallel to the page, in good light, without shadows across the words.
• Screenshots read best, because the letters are perfectly sharp. A screenshot at your screen’s full size beats a zoomed-out one.
• Dark text on a light background reads more reliably than text over a photo or a pattern.

What comes out

One .txt file per image, named after it, with the text in reading order: columns are read one after another, and each paragraph keeps its own lines. Tables come out as lines of text rather than cells. Very large photos are scaled down to about 16 megapixels first, which is still far more detail than text needs.

Private by design

The reading runs in your browser, in a separate worker that can only fetch files from this site, so the picture never leaves your device. The engine is Tesseract, the open-source OCR engine, built to run in the browser. Once the page has loaded it, the tool keeps working even if your connection drops.

Good to know

• For a scanned PDF, use PDF OCR, which reads every page.
• For a PDF that already has text you can select, PDF to Text is faster and exact.
• A wrong word or two is normal in OCR; read the result through before you send it on.