Feature: Support of Video/Audio -> Text?

It is great that the current design solve the problem of real time ASR. However, I can see a bigger potential and a fix workaround towards static file, e.g. Video or Audio -> Text via ffmpeg wrapper.

This way not only we can record our voice -> Text to extract transcript, but also allow us to upload old audio or video file to have transcript as MacOS native solution.

No files or tests are named. Start by reviewing the existing real-time ASR flow and how the app handles audio or video input, then assess the proposed ffmpeg-wrapper approach for static files. Done means the upload, supported formats, transcription flow, and acceptance criteria are defined.

Written by the indexing model from the issue text.