Image to Text (OCR)
Free OCR tool to extract text from an image, screenshot or scanned PDF. Recognise the text in a picture in your browser — in a dozen languages — then copy, edit or download it as a .txt file. No upload, no watermark, no signup.
🔒 100% private — the text is recognised in your browser and never uploaded. The first run loads the OCR engine (~3 MB) plus the language data, then it’s cached.
Extract text from an image or PDF — free, no upload
This free OCR (optical character recognition) tool pulls the text out of a picture, screenshot or scanned PDF and turns it into text you can copy, edit and download. Everything runs in your browser — no upload, no watermark, no signup.
What can it read?
Photos of documents, screenshots, scanned pages, receipts, slides and PDFs. For best results use a sharp, well-lit image where the text is roughly upright. You can pick from a dozen languages, including Chinese, Japanese, Arabic and Russian.
Is it private?
Yes. The recognition engine (Tesseract, compiled to WebAssembly) runs entirely on your device, so your image or PDF never leaves your computer. Paid OCR sites upload your files — this one doesn’t.
How do I get the most accurate result?
Use the highest-resolution image you have, make sure the text isn’t skewed or blurry, and choose the correct language. You can also fix any stray characters by editing the text box before you copy or download it.
Frequently asked questions
Is the OCR really free and private?⌄
Yes. The text recognition runs entirely in your browser using Tesseract compiled to WebAssembly, so your image or PDF is never uploaded and there's no signup or paywall — unlike paid OCR sites that upload your files.
Can it read a scanned PDF?⌄
Yes. Each PDF page is rendered to an image and read in turn (up to 30 pages), and the text is combined into one result you can copy or download.
Which languages does it support?⌄
English, Spanish, French, German, Italian, Portuguese, Dutch, Chinese (Simplified), Japanese, Arabic, Russian and Hindi. Pick the language of the text before you run it for the best accuracy.
Why does it download something the first time?⌄
The first run downloads the OCR engine (about 3 MB) plus the language data into your browser and caches it. After that, recognition starts straight away.
How do I get the most accurate result?⌄
Use a sharp, high-resolution image where the text is upright and well-lit, and choose the correct language. You can fix any stray characters by editing the text box before copying or downloading.
Have another question? Get in touch.