Cheese! OCR

Select any screen area, get text instantly

Mac

Productivity

Privacy

GitHub

Hey Product Hunt! 👋 I built Cheese! OCR because I kept running into the same frustration during my daily workflow — and I bet many of you have too. When I'm writing academic papers, I constantly need to reference text from PDF files that don't allow copy-paste (scanned documents, locked PDFs, image-based charts with labels). I used to retype everything by hand or rely on cloud-based OCR tools that required uploading my research materials to some server. Neither felt right. Another scenario that happens all the time: grabbing error messages or log outputs from a video tutorial or a screenshot someone shared in Slack. You can't select text in an image — but with Cheese! OCR, you just press a hotkey, drag over the area, and the text is in your clipboard instantly. So I built something simple: a menu bar app that does OCR entirely on-device using Apple's Vision framework. No uploads, no network requests, no accounts — the app doesn't even have network permissions. Your screenshots never leave your Mac. It supports English, Chinese, Japanese, and Korean out of the box, which is essential for multilingual researchers and developers. The name? Just like saying "Cheese!" when taking a photo — our little document mascot smiles every time you capture text :) Would love to hear your feedback!