Building CaptionSpark: live captions for audio inside Chrome

I recently shipped CaptionSpark, a Chrome extension that turns browser audio into live captions.

The narrow use case is this: you are already on a page, audio is playing, and you want readable text without opening another app. That can be a video, a webinar, an X Space, a call, or a stream.

The extension starts from the active tab and puts captions over the current page. It also supports Spanish translation and keeps a transcript while the session is running.

What I learned while building it:

1. Caption UX is mostly about trust. People need to know when capture is active.


2. Latency matters more than perfect layout. If text arrives late, the overlay feels broken.


3. The overlay has to be visible but not loud.


4. Pricing by minutes is clearer than pretending transcription has no cost.

It is early, but working. I would appreciate feedback on the onboarding and the overlay.

Chrome Web Store:

https://chromewebstore.google.com/detail/captionspark-browser-audi/phmkpkbjecdlklhmemgncabkddbpapnn