Stop the copy-paste loop. ScreenTranslate is a lightweight macOS utility for instant on-screen translation. Inspired by ScreenHint’s seamless capture, I built this to solve a personal pain: translating uncopyable text in images, video frames, and locked PDFs. 1)Drag-to-Translate: Instant results by dragging over any area. 2)Stay in Flow: No more switching to Google Translate tabs. 3) Privacy-First: Native macOS experience with no data collection.
immersivetranslate
ScreenTranslate
ScreenTranslate
@vouchy Thanks for the warm welcome! To answer your question—yes, there was a very specific 'aha' moment.
I’ve always been sensitive to tiny frictions that break my flow. The tipping point was when I was diving into the source code of OpenClaw to understand how AI agents structured their system prompts. I found myself stuck in a tedious loop: 'Copy raw text → Switch to Chrome → Search Google Translate → Paste & Wait → Read → Switch back to VS Code.' Doing this dozens of times a day felt like a massive tax on my productivity. The most important goal for me was the UX—I wanted to understand the text instantly without the active window ever changing. I’m a heavy user of ScreenHint, and its drag-to-action interaction gave me the perfect hint for how ScreenTranslate should work.
I actually used ScreenTranslate to read and reply to your comment just now! 😊 It’s becoming an essential tool for me, especially for parsing long English responses from Gemini or Claude quickly. For me, the ultimate goal of AI is exactly this: building personal tools that solve my own frustrations and keep me in 'the zone.' Thanks again for noticing!
It does the same with audio files, I saw videos being mentioned in the description, how does that work? Kudos on having it as a pop-up so that we do not have to jump between tabs, that takes some frustration away
ScreenTranslate
@viktorgems Hi Victor! Thanks for the kudos on the pop-up UX—I’m thrilled to hear that it’s helping you stay in the flow, as that was exactly my goal!
Regarding your question about video and audio: I apologize for any confusion in the description. Currently, ScreenTranslate focuses on translating the 'visual text' within the video screen (like subtitles or on-screen captions) rather than translating the audio or speech itself.
It works by capturing the text appearing on the screen in real-time. I’ll make sure to clarify this in the description to avoid further misunderstanding. I really appreciate you pointing that out, and thanks again for the support!
That’s cool Changmin congrats on your 1st launch 👏
Quick question
How many languages does it currently support?
Also, any plans to release a version for Windows users?
ScreenTranslate
@jivitharsan_suresh Thank you so much for the kind words and great questions! 🙏
Languages: ScreenTranslate currently supports 20 languages — Arabic, Chinese (Simplified & Traditional), Dutch, English, French, German, Hindi, Indonesian,Italian, Japanese, Korean, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Thai, Turkish, Ukrainian, and Vietnamese — all powered by Apple's built-in Translation framework, processed entirely on-device.
I actually just updated the website with a Supported Languages section since I realized this info was missing.
Apologies for the oversight — This is my first launch and I'm still learning!
I'm also planning to add optional third-party OCR engines so users can unlock even more languages beyond the current 20 (which is limited by Apple's built-in OCR).
Windows: I'd love to bring ScreenTranslate to Windows someday! However, one of the core values of this app is 100% on-device processing — and that's possible on macOS thanks to Apple's built-in Translation and Vision frameworks. Windows unfortunately doesn't have an equivalent system-level on-device translation API yet, so achieving the same privacy-first, offline experience would require a fundamentally different approach. It's definitely something I'm keeping an eye on, but for now I'm focused on making the macOS version the best it can be.
Thanks again for the feedback!
ScreenTranslate
@alamenigma
Thank you so much for the kind words, Modassir!
Great question — I've actually already added support for multiple translation engines! In the latest update, you can bring your own API key (BYOK) for DeepL, Google Cloud, and Microsoft Azure alongside the built-in Apple Translation. So you can choose whichever engine works best for your language pair. (https://screentranslate.filient.ai/engines)
Here's how it currently breaks down:
- Screen capture translation (drag-to-translate via OCR): Currently powered by Apple Vision OCR, which supports 20 languages. The OCR recognition is the bottleneck here, not the translation engine itself.
- Text selection translation (select text + shortcut, no OCR needed): Works with all supported languages of your chosen translation engine — so DeepL, Google Cloud, or Azure can cover many more languages beyond the 20 OCR-supported ones.
For expanding OCR language coverage, I'm planning to add Tesseract and Surya as optional downloadable OCR engines. The idea is to keep the base app lightweight while letting users install additional OCR engines only if they need broader language support. This way, privacy-first users can stay fully local, and power users can extend as needed.
Thanks again for the support — feedback like this really keeps me motivated!