What it does
Live speech mode requires macOS 26 plus camera and microphone access; typed input can drive the effects without that hardware.
Apps & data pipelines
jev-telop-live
Turn live Japanese speech into variety-show captions, manga effects, colored lenses, and sound cues.
Bookmark: jev-telop-live
Keep this in your collection.
Leave a noteWhat would you try with this?
: jev-telop-live
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
A developer can adapt this project to add live visual effects to a video stream. The source code shows how to take spoken Japanese and send it to an outside AI service called TypeSafe. TypeSafe answers multiple questions at once to choose matching sound effects, mood changes, and large captions.
To try the live speech feature, your developer needs a Mac running macOS 26. They also need an access key that connects the app to TypeSafe. You must wear headphones so the microphone ignores the computer's sound effects. Without a microphone, you can type text to trigger the visuals manually.