Super cool! Did you end up doing this as 2 separate jobs and then chaining them together? (i.e. DVM #1 transcribes the video then DVM #2 summarizes the transcription)
How are you expecting to post the final DVM result as a note? I should be able to write up a bot that does all this in one step if you think it would help.
By this I mean:
A video gets posted nostr
User summons bot by commenting on video
Bot replies with fee to transcribe and summarize
User pays invoice
Bot replies with summary
Let me know if this would be valuable to you or if you have something else in mind. Cool idea!