Choose an image or paste a screenshot, then click Extract text; the recognized text appears below once the scan finishes.
Worked example
Raw OCR output keeps the line breaks of the original image, so words get split by hyphens and sentences get chopped mid-line. This is what the scan of a memo snippet returns:
One click of Tidy text joins the split words and unwraps the lines - without touching spelling:
Lines that end in punctuation or start a new list item keep their line breaks, so numbered lists and headings survive the tidy pass unchanged.
Image to Text (OCR)
Pull the text out of a photo, screenshot, or scanned page without installing anything or sending the image anywhere. The image opens in this browser tab, the recognition engine runs locally, and the recognized text lands in a box you can copy or download as a .txt file - the file itself never leaves your device.
Key features
- Opens a local image file in this browser tab - it is never uploaded to a server.
- Takes a screenshot straight from the clipboard: press Ctrl+V (Cmd+V on a Mac) on the page, with no need to save the screenshot as a file first.
- Runs the recognition engine locally, so the image and the recognized text never leave your device.
- Drops the recognized text into an editable box you can copy to the clipboard.
- Downloads the result as a .txt file built in the browser, with no server round-trip.
When to use it
Works best on clear, printed text with good contrast - accuracy drops on handwriting, low-resolution photos, or heavily skewed or rotated text.
Reading words out of an image is a different job from describing the image itself. For handwritten notes, the handwriting to text converter runs a handwriting-trained model on your device and handles that job better. If what you need is a caption - a one-sentence description of the scene for an alt attribute - the alt text generator drafts one per image on your device, with the same no-upload rule as this page.
Recognition covers one language at a time (English by default) and returns plain text only - it does not preserve the original layout, columns, or fonts, and does not return bounding-box coordinates for the recognized words.
Cleaning up the recognized text
Raw OCR output mirrors the line breaks of the image, so words arrive split by hyphens and sentences arrive chopped mid-line. The Tidy text button fixes exactly that - it joins the split words, unwraps hard-wrapped lines, and collapses repeated spaces, and it never changes spelling. In browsers that ship a built-in proofreading model, a Fix errors with AI button also appears; it corrects likely misreads on your device, but it can change words that were already correct, so review its result - an Undo button restores the raw text after either cleanup.
Privacy and first run
The image itself is never uploaded. The first run on a device downloads the recognition engine and language data (a few megabytes) before it can scan anything; the browser caches that download so later runs on the same device start faster.
Frequently Asked Questions
Does the image get uploaded anywhere?
No. The image opens in this browser tab and the recognition runs locally - it is never sent to a server. The only network activity is a one-time download of the recognition engine and language data on the first run on a device (a few megabytes); later runs on that device are faster because the browser caches it.
Does it work on handwriting?
Not reliably. The recognition engine is trained on printed text, so results are best on clear, printed text with good contrast. Accuracy drops on handwriting, low-resolution photos, and heavily skewed or rotated text.
Does it keep the original layout, like columns or tables?
No. The output is plain recognized text only - it does not preserve the original layout, columns, tables, or fonts, and does not return bounding-box coordinates for the recognized words.
Can it recognize more than one language at once?
It recognizes one language at a time (English by default) and does not auto-detect the language in the image, so mixed-language images will misrecognize the portions in the non-selected language.
Can I paste a screenshot instead of choosing a file?
Yes. Press Ctrl+V (Cmd+V on a Mac) anywhere on the page and the screenshot on your clipboard loads straight into the tool - no need to save it as a file first. The Paste screenshot button does the same thing, though some browsers ask for clipboard permission before it works; the Ctrl+V shortcut needs no permission.
What do Tidy text and Fix errors with AI do - and can they be wrong?
Tidy text is deterministic: it joins words the scan split across line breaks, unwraps hard-wrapped lines, and collapses repeated spaces - it never changes spelling, and lines ending in punctuation keep their breaks so lists survive. Fix errors with AI appears only in browsers that ship a built-in proofreading model; it corrects likely misreads on your device, but it can also change words that were recognized correctly, so review its output. Undo last cleanup restores the raw text after either one.