Free OCR for printed text, entirely in your browser
Drop an image, get the text back. Nothing is uploaded anywhere - the recognition runs on your own device.
This tool is for printed text - book pages, receipts, screenshots, signs. It is not built for handwriting and will do poorly on it.
Drag & drop an image here
or click to browse, or paste with Ctrl+V (Cmd+V on Mac) - JPG, PNG, WebP, or BMP
Why this stays entirely in your browser
The recognition engine (an open-source OCR model called Tesseract, compiled to WebAssembly) downloads once and then runs on your own device's CPU. Your image is never sent to Scriveno or anyone else - there's no upload step at all, which also means no account and no limit on how many images you process.
When you need more than plain text
This tool is deliberately simple: it reads text, nothing else. If your page has equations, chemical structures, or diagrams mixed in, plain OCR will mangle them - that needs a model trained specifically on math and science notation, plus a way to keep diagrams as diagrams instead of garbled text. That's what Scriveno's main converter does: upload a photo of handwritten math, physics, or chemistry notes and get back real LaTeX, a typeset PDF, an editable DOCX, and Markdown - free to start.
FAQ
Does this work on scanned PDFs?
Not yet - this tool accepts a single image (JPG, PNG, WebP, or BMP). For a multi-page PDF, convert each page to an image first, or use Scriveno's main converter, which accepts PDFs directly.
Why is it slower than a cloud OCR service?
It's running on your device's CPU instead of a data center's - the tradeoff for your image never leaving your machine. A typical page takes a few seconds on most laptops.