Minitooler

Extract text from images and scanned PDFs (OCR)

Run OCR on images and scanned PDFs entirely in your browser — extract text from JPG, PNG, WebP, and PDF in 20+ languages. No upload, no signup.

Loading tool…

What people use it for

  • Turn a screenshot of a receipt, invoice, or slide into editable text.
  • Extract quotes from a scanned book page or newspaper clipping.
  • Copy text out of a photo of a whiteboard or handwritten sign (printed lettering).
  • Convert scanned contracts and legal PDFs into searchable text.
  • Digitize old research papers, letters, or handwritten notes (printed only).
  • Pull data out of screenshots you can't select text from.
  • Feed printed content into ChatGPT, Notion, or a spreadsheet.

A free online OCR tool that runs 100% in your browser — drop an image to text or a scanned PDF to text and get editable output in seconds. No signup, no email, no upload. Whether you need to extract text from an image, convert a picture to text, or turn a scanned PDF into searchable text, this tool does it locally with a text-recognition engine that runs right on your device.

Great for turning screenshots into text, digitizing scanned documents, pulling copy off a photo, or making an old image-only PDF searchable. Because everything happens in your browser, it is a safe way to run OCR on confidential documents — contracts, invoices, medical records, IDs, tax forms — without trusting a random web service with your data.

Supports JPG, PNG, WebP, and PDF in over 20 languages including English, Spanish, French, German, Italian, Portuguese, Danish, Russian, Arabic, Hebrew, Hindi, Japanese, Korean, and Chinese. Pick the right language before you scan — recognition is much more accurate when it knows what alphabet to expect.

How it works

  1. 1

    Pick a language

    Choose the language of the text in your file — English is default. The tool downloads that language pack once and caches it.

  2. 2

    Drop your file

    Drag in an image or a scanned PDF. Nothing is uploaded — recognition happens on your device.

  3. 3

    Copy or download

    Grab the recognized text via clipboard or download it as a .txt file.

Frequently asked questions

Is my image or PDF uploaded anywhere?
No. OCR runs entirely in your browser with a text-recognition engine that runs on your device. Nothing about your file — pixels, text, or filename — is sent to a server.
Which languages are supported?
20+ languages including English, Spanish, French, German, Italian, Portuguese, Dutch, Danish, Swedish, Norwegian, Finnish, Polish, Russian, Ukrainian, Turkish, Arabic, Hebrew, Hindi, Japanese, Korean, and Chinese (Simplified and Traditional). Pick the language first for the best accuracy.
Does it work on scanned PDFs?
Yes. Each page is rendered to an image in your browser, then OCR'd page by page. If your PDF already contains real text, use the PDF Text Extractor instead — it's instant and lossless.
What about handwriting?
The engine is trained on printed text, not handwriting. Neat block letters may work, cursive rarely does. For printed pages, screenshots, invoices, and receipts it's very accurate.
Why does the first run take a while?
The language model (~10–20 MB) downloads once, then is cached. After that, recognition runs offline and quickly.
What's the file size limit?
Free users can process files up to 10 MB. Pro users are only limited by their device's memory.
How many files can I OCR per day?
Guests get 3 free runs a day across all tools, and a free account raises that to 5 in total. Pro users ($5/month) get unlimited use.