There’s a reason it’s being implemented in an open source application that gets actual use instead of a random-word-generated-elevator-pitch being chased by VC funds and dinosaurs in the stock market.

In fact it's a feature I've been waiting for someone to do!

Windows has this as a feature in the OS with their stupidly-named Copilot+ PCs. I tried it a while ago and it seemed ok-ish at translating. Wasn’t terrible for doing it all on-device

https://support.microsoft.com/en-us/windows/use-live-caption...

Finally something that can potentially run on those new shoehorned NPUs and isn't just more spying.