Extract text from images and PDFs (OCR)
Copy the text out of photos, screenshots and scanned PDFs in English, Hindi and seven other languages. Your file is read on your device.
Runs in your browser. Your file never leaves this device.
Loading the tool…
What this tool does
Pull the text out of a photo, a screenshot or a scanned PDF so you can copy, edit or search it. That saves you retyping a printed letter, a notice, a page from a textbook or a scanned form.
The text recognition is done by Tesseract, a well-known open-source OCR engine, running in your browser. Your file is never sent anywhere. The first time you use it, your browser downloads the engine and the language data from jsDelivr, a public code server. The language data is then stored in your browser for next time.
You can pick one language at a time: English, Hindi, Spanish, French, German, Italian, Portuguese, Simplified Chinese or Japanese. For a PDF, every page is read in turn and the text is joined with a marker between pages.
You get plain text that you can edit in the box, copy, or download as a .txt file. It doesn't make a searchable PDF. Results are best with clear, printed text. Handwriting, blurry phone photos and decorative fonts give patchy results, so always read the text through before you use it.
People use it for
- Copying the text of a printed notice or letter instead of retyping it
- Getting editable text from a scanned page of a textbook or notes
- Pulling details from a screenshot of a bill or receipt
- Reading text from a scanned PDF so you can search it or paste it into a document
How to use it
Add an image or PDF
Drop a photo, screenshot or scanned PDF onto the box.
Choose the language
Pick the language the text is written in. English is selected to start with. Choosing the right one makes a big difference to accuracy.
Read, copy or download
Recognition runs page by page. Fix any mistakes in the text box, then copy the text or download it as a .txt file.
What it works with
| You give it | You get | Good to know |
|---|---|---|
| JPG / PNG / WEBP | Plain text | Clear, straight photos work best |
| Plain text | Every page read; text joined with page markers |
Questions
Which languages are supported?
English, Hindi, Spanish, French, German, Italian, Portuguese, Simplified Chinese and Japanese, one at a time. A page mixing Hindi and English will only be read well in the language you pick.
How accurate is it?
Very good on clean, printed text. Accuracy drops with handwriting, low light, tilted photos, small print and decorative fonts. Check the text before relying on it.
Can it make my PDF searchable?
No. It gives you plain text to copy or save as a .txt file. The original PDF isn't changed.
How do I get better results?
Take the photo straight on, in good light, with the page filling the frame. Crop out anything that isn't text, and pick the correct language.
Why is it slow the first time?
Your browser downloads the OCR engine and the language data before it can start. The language data is saved in your browser, so the next run is faster.
Is my document uploaded?
No. Recognition happens on your device. Only the engine and language data are downloaded, from jsDelivr; your file isn't sent.