Please improve transcription accuracy asap

Hey team,

Love Screen Studio for recording demos — the auto-zoom, smooth cursor, and caption styling are all fantastic. The word-by-word highlighting looks really polished and professional.

The one thing holding me back dramatically is transcription accuracy. Even on the Medium model, the Whisper transcriptions have enough errors that I end up spending a lot of time in the transcript editor fixing things. Plus editing is horrible and messes up the timing of the transcripts and the frames.

It would be a huge improvement if 1. the transcript quality and edibility gets upgraded or 2. Screen Studio supported importing an SRT or VTT file. That way I could use a more accurate external transcription service and still get the beautiful caption styling that Screen Studio does so well. Right now there's no way to bring in a corrected transcript from outside, which means I'm stuck choosing between Screen Studio's great subtitle formatting and actually accurate subtitles.

This would be a game-changer for anyone making polished product demos or tutorials where subtitle quality really matters.

Thanks for building a great product, hope this is on the roadmap!