Is this OCR tool free?

Yes. Free, unlimited use, no signup. Runs in your browser using Tesseract.js.

Are my images uploaded anywhere?

No. OCR runs entirely in your browser. The image and extracted text never reach our servers.

How accurate is the OCR?

Very accurate on clean printed text (95%+ on screenshots, typed documents, scanned books). Lower accuracy on handwriting, low-resolution captures, or heavily skewed images. For best results, use clear, high-contrast images with at least 300x300 px resolution.

Which languages are supported?

100+ languages including English, Spanish, French, German, Italian, Portuguese, Chinese (simplified and traditional), Japanese, Korean, Arabic, Russian, Hindi, and many more. Pick the language that matches the text in your image.

Why is the first run slow?

The OCR engine and language data (~3 MB) downloads on the first extraction. After that, it is cached and subsequent extractions are fast.

Can it read handwriting?

Tesseract.js handles cursive and handwriting poorly. It is optimised for printed and typeset text. For handwriting, the accuracy is often below 50%.

Not directly. First convert the PDF to JPG using our PDF to JPG converter, then OCR each page.

Image to Text (OCR) - Free Online OCR, Extract Text from Photos

Drop image here or click to upload

JPG, PNG, WEBP - paste from clipboard works too

Cmd+V to paste a screenshot

Free OCR Online - Image to Text in Your Browser

Optical Character Recognition (OCR) reads printed or typeset text out of an image and converts it into editable, searchable text. This tool runs OCR entirely in your browser using Tesseract.js, an open-source engine maintained by Google. Your images never reach any server.

The first time you extract text, the engine and language data (~3 MB) download to your browser cache. Every subsequent run is fast because the engine is already loaded.

How to Get the Best OCR Results

Use high-resolution images. At least 300x300 pixels for short text, 1000+ pixels for paragraphs. Higher resolution gives the engine more pixels to work with.
Make sure the text is horizontal. Skewed or rotated text degrades accuracy. Use our Rotate tool first if needed.
Prefer high contrast. Black text on white background is ideal. Faded scans benefit from a quick contrast boost in any photo editor.
Pick the right language. Tesseract is much more accurate when you tell it which language to expect.
Crop tightly. Less surrounding noise means less for the engine to misinterpret. Use our Image Crop tool to trim.

What OCR Is Good At - And What It Is Not

Source	Expected accuracy
Screenshots of webpages	95-99%
Typed documents (clean scans)	95-99%
Photographed book pages (good light)	90-97%
Phone photos of receipts	70-90%
Handwritten text	30-60%
Cursive handwriting	10-30%
Faded or low-contrast scans	40-70%

Common Use Cases

Convert a screenshot of an article into editable text.
Pull text out of a meme, infographic, or shared image.
Digitise printed receipts, contracts, or notes.
Make a scanned PDF page searchable (use our PDF to JPG converter first).
Translate text in a foreign-language image (extract here, then paste into a translator).