Uber Tools

Image to Text

Copy the text out of photos, screenshots and scans (OCR), read in your browser, nothing is uploaded

Up to 3, each model is downloaded once (1–3 MB)

Layout

Drop images with text here

Or choose them, or paste a screenshot (Ctrl / ⌘ + V). PNG, JPEG, WebP, GIF, BMP

Works best on sharp, straight, printed text with good contrast; handwriting is mostly not read. A language's model is downloaded from this site the first time it's used and kept by your browser, the images never leave your device.


How to extract text from an image

  1. Pick the languages in the image, up to 3.
  2. Drop your images on the page, choose them, or paste a screenshot (Ctrl or ⌘ + V): PNG, JPEG, WebP, GIF or BMP.
  3. Choose the layout if needed: automatic, one block, one line or scattered text.
  4. Copy the text or download it; words read with low confidence are marked on the image.

Questions

Can it read handwriting?

Mostly not: it works best on sharp, straight, printed text with good contrast, such as screenshots, scans and photos of documents.

Which languages can it read?

Many: pick up to three per image. Each language's model (1 to 3 MB) is downloaded from this site the first time and kept by your browser.

Are my images uploaded?

No. The text is read by your browser, on your device (with Tesseract); only the language models are downloaded.

How do I read a scanned PDF?

Turn its pages into images with PDF to JPG, then read them here.

Related tools