Improved Audio Transcription

Hey team! I think the new audio transcription feature is awesome!

There are tons of tools like Granola, Notion AI, etc. that companies use to take meeting notes. But often, you want to transcribe meetings without having those recordings stored somewhere your company's IT team can access.

The current feature is already helpful, but I think a few simple additions could make it 10x better:

1. Speaker identification. Transcripts should show who's talking. Ideally, AI could figure out names from the conversation itself, since people often mention each other's names. It could also recognize your voice through your mic. But even basic labels like Speaker 1, Speaker 2, etc. would be a huge improvement. It's really hard to follow a transcript without knowing who's saying what.
2. Automatic meeting detection. It would be great if Capacities could detect when a meeting starts and show a popup asking if you want to start transcribing.
3. Live transcript popup - I'd love a small, semi-transparent window showing the live transcript. You could quickly scroll back to see what people just said if you missed something or need to catch up.

Overall, I think these additions could make this my favorite feature in Capacities!

Capacities v1.72.4; none none; Mac OS v10.15.7; Chrome v148.0.7778.280 (WebKit v537.36); Desktop (Electron); 10/27/2026, 1:21:12 PM