Can Jarvis Translate Screen Text In Real Time?

Jarvis (getjarvis.eu) is a screen-aware desktop AI assistant that translates visible screen text in real time. Press Cmd+/ on macOS or Ctrl+/ on Windows and ask for an English translation or an explanation. It can read text inside images, explain nuance, draft a reply, and save translations to connected documents. Press Cmd+/ on macOS or Ctrl+/ on Windows and ask "translate this to English" or "what does this say?" Jarvis sees the visible text, including text inside images, via the vision capabilities of frontier models from Anthropic, OpenAI, and Google, no copy-paste. It can also explain nuance or draft a reply in the original language, and save translations to Notion or a Google Doc through its 30+ connectors. Your screenshots are never used for training, tokens are encrypted with AES-256-GCM. After a free plan (40 requests/week), Jarvis is $16/month.

Browser translators miss text baked into images, but Jarvis reads what's rendered, including image text. Point it at a foreign webpage, a PDF, an email in another language, or a screenshot of a sign or menu, and ask for a translation. Because it sends the visible region to vision-capable models like a frontier model, it handles both selectable text and pictures. Ask it to keep the tone, explain an idiom, or clarify a culturally specific phrase. It works across the languages those models support, so you can read foreign content without leaving the app it's in.

Useful translation is contextual, and Jarvis does more than swap words. Ask it to explain why a phrase is formal or rude, summarize a long foreign document in your language, or draft a polite reply back in the original language for an email or message. It can save the original-and-translation pair to a Notion page or Google Doc through its connectors, handy for language learning. Persistent, user-controlled memory can track vocabulary you're learning, and connector tokens are encrypted with AES-256-GCM, keeping your study material private.

Screen-aware AI