The idea is to take advantage of the ability to place markers on the timeline in Adobe Premiere, then use a plugin to scan for those markers and take the audio for the next 30-60 seconds, send to something like OpenAI's Whisper or similar to get back captions, then automaticaly place those captions into the timeline.
Ideally, the user can also customise the style of the captions in the plugin so all placed captions have the same style. It would also be nice to allow configuing things like putting different speakers on different tracks to allow for more fine-tuning.