I know your focus is on the podcast/audio side of things (and you nailed this, very very cool), but it would be interesting to me to see how you generated the s

I know your focus is on the podcast/audio side of things (and you nailed this, very very cool), but it would be interesting to me to see how you generated the summaries themselves, if that was available, or at least a brief summary of the code, to understand what it took to get to that point.