By skyforbes Currently, we have ~400 hours of podcast interviews with leading experts in our field. All ~250 transcripts have been collected. We would like to most effectively use our podcast to train our custom GPT.
How should this vast, static information be structured in a markdown file(s)? Is there a good way to keep this knowledge up-to-date in the future, probably by topic, so that it can be referenced as best as possible?
PS. There is no need to index the podcasts, guests, times, lengths, etc…The GPT should absorb as much knowledge as possible about our field, current events, etc.
Thanks for any help!
