I'm intrigued by the prospect of AI learning from our collection of videos and that informing something about the edits to future videos.
Q 1 : Can that be segmented? Like, I'll be editing videos of 3 different clients, who's styles are very different.
Q2: If it's "Learning" could we teach it up-front, before it processes the audio?
Example A: , my client is company named "Kosmic Ai"...
Can we pre-feed a style guide to the AI, such that whenever "Cosmic A.I." is spoken, it would transcribe that as "Kosmic Ai", as that is the actual brand name for this client? (AND that it should never be split into two captions?)
Example B: I'm currently editing videos about jazz theory, with demos.
I'd love to pre-teach Kosmic Ai that:
G Sharp == write as ==> G#
D over G == write as ==> D/G.
(and never split those either).