Apple quietly shipped everything you need to build a real-time translator — so I built one
I work at a German company. Meetings are in German, sometimes English — and even though I speak both, there are moments in fast meetings where I zone out for two seconds and think: wait, what did they just say? I couldn't use any of the existing captioning tools, because they all pipe your meeting audio to a cloud server. Sending confidential work calls to a third party was a non-starter. Then I…
A German journalist and software developer with expertise in macOS frameworks built a real-time translator app called Wakaru. Existing captioning tools send audio to cloud servers, which the journalist didn't want for confidential work calls. macOS 26 included the necessary building blocks for on-device speech recognition, machine translation, and large language models, enabling the app to run locally without servers or API keys.
The app captures system audio, analyzes speech, translates it, and overlays subtitles directly on the screen. The journalist emphasizes the importance of waiting for complete sentences before translating, setting the correct audio sample rate, and remembering the last few words of translated sentences to avoid rewriting history.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
