So Ive built a pretty extensive Social Media posting automation, it know past post as not to repeat, it know which past media was posted not to repeat, agent creates the titles, descriptions, hashtags.
This has been done over months of testing with real world clients and has been in production for over 6 months, but I want to step it up and make that final push to make it what I believe will be a complete system.
Right now it uses information about the company to come up with titles, descriptions and such.
So IM here to ask all you what LLM can I use that can watch the video, have context of that video so it can be used to make even more relevant posts? Also any suggestion on how to go about this. It must watch the video and have context as almost all the video have no dialog so it not getting any context just from a simple transcribe.
Thanks everyone!