Screenshot to Text — Free OCR

Extract text from any screenshot instantly — error messages, code, menus, dialogs, or phone captures. Runs Tesseract OCR in your browser. No upload, no account, 19+ languages. Works on Windows, Mac, iPhone, and Android screenshots.

100% Private — No Uploads

Copy to Clipboard Instantly

19+ Languages · Bulk Up to 50

Windows ScreenshotMac ScreenshotiPhoneAndroidWebPPNGJPG

Drop screenshots, photos, or scans here or click to upload

.JPG · .PNG · .WebP · .BMP · .GIF · .TIFF · up to 50 files per batch

OCR runs in your browser — nothing uploaded

Tesseract.js (a WebAssembly build of the Tesseract OCR engine) executes locally on the page. Your images and the extracted text never leave your device. Sensitive scans — receipts, IDs, medical reports, contracts — stay private.

How to extract text from a screenshot

1
Take or open your screenshot
Use your screenshot (PNG from Windows/Mac, JPG from iPhone/Android, or any image of a screen). Drop it into the upload area or click to select it.
2
Select language if needed
The default is English. If your screenshot contains text in another language (Spanish, French, German, Arabic, Japanese, etc.), pick it from the dropdown. The language pack downloads once and is cached.
3
Click Extract
Tesseract.js runs OCR in your browser. For a typical screenshot this takes 2–10 seconds. A progress indicator shows per-file status.
4
Copy or download the text
Click Copy to clipboard and paste wherever you need it. Or download as a .txt file. For multiple screenshots, download all as a ZIP.

When would you need to extract text from a screenshot?

Copy an error message from a crash dialog to search for a fix
Extract code from a tutorial video where you can't copy-paste from the source
Copy a street address from a screenshot of an email or map
Grab product names, prices, or tracking numbers from a screenshot of a receipt or order
Copy subtitles or captions from a video screenshot
Extract text from a screenshot of a PDF that won't let you select text
Digitize notes you photographed on a whiteboard or chalkboard

Why screenshots give the best OCR accuracy

Compared to photos of printed documents, screenshots are ideal for OCR: they are always high-DPI (typically 1x–3x), have perfect contrast (black text on white or near-white background), and use clean system fonts (San Francisco, Segoe UI, Roboto) that Tesseract recognises reliably. In practice, Tesseract achieves 98%+ character accuracy on typical UI screenshots — error messages, dialog boxes, terminal output, and menu items all extract cleanly in most cases.

Tips for better results on tricky screenshots

Crop to the text area — removing surrounding chrome (toolbars, wallpaper) improves accuracy and speed
Avoid dark mode screenshots — light text on dark background can confuse Tesseract; invert colours first if needed
Match the language — picking the wrong language pack causes substitution errors, especially on non-Latin scripts
High DPI for tiny text — if text is very small, take a zoomed-in screenshot rather than cropping and upscaling

Frequently asked questions

How do I copy text from a screenshot?

Drop your screenshot into this page, click Extract, and copy the result to clipboard. The OCR runs locally in your browser using Tesseract WebAssembly — no upload, no account needed. Works with PNG, JPG, and WebP screenshots from Windows, Mac, iPhone, and Android.

Is my screenshot uploaded to a server?

No. Tesseract.js runs as a WebAssembly module directly inside your browser tab. Your screenshot stays in RAM on your device. Nothing is sent to any server — not the image, not the extracted text. You can disconnect from the internet after the page loads and the tool still works.

What screenshot formats are supported?

PNG (the default format for Windows and Mac screenshots), JPG, WebP, BMP, GIF, and TIFF. Phone screenshots are usually JPEG or PNG — both work fine.

How accurate is the OCR on screenshots?

Very accurate on typical screenshots — interface text, menu items, error messages, dialog boxes, and code. Screenshots usually have high DPI, high contrast, and clean system fonts, which are ideal conditions for Tesseract. Handwriting and decorative fonts are harder.

Can I extract text from a phone screenshot?

Yes. Open this page on your phone, tap to select your screenshot from the gallery, and tap Extract. The OCR runs on the phone's CPU in the browser. Alternatively, transfer the screenshot to your desktop and run it there — desktop CPUs process the OCR faster.

Does it work with screenshots of code or terminal output?

Yes, but expect occasional misreads on special characters like brackets, underscores, and pipes. Tesseract is trained on natural-language text. For best results, crop tightly around the code block and use a screenshot with a plain background.

Can I extract text from multiple screenshots at once?

Yes — drop up to 50 screenshots at once and click Extract All. Each is processed in sequence and you can copy or download each result individually, or download all as a ZIP of .txt files.