TheOneWhoKnows Posted April 29 Posted April 29 I am going to be a human grenade and just drop this here and watch the aftermath…. Quote
NewArpege Posted April 29 Posted April 29 Interesting experiment. It’s hard to judge off videos, imo. So many times I’m not a fan of a show on video, but then blow away in person. But I love seeing people experiment and push boundaries. 1 Quote
KVG_DC Posted April 29 Posted April 29 Haaaaaaa! The whole time I was like "wait...WHERE is Phantom Regiment?" Also BD seemed super hype for where they placed with that editing. The long and short of it though is...I don't think AI can figure out what GE is either. 5 Quote
KVG_DC Posted April 29 Posted April 29 (edited) Dude asked it to give reasoning for its scores too. https://docs.google.com/spreadsheets/d/1ICQD4pWp9hVhdFHpK_clxzVnCrkEM43p/edit?gid=1751986502#gid=1751986502 AI is jargon-ing so hard with so much confidence here... Edited April 29 by KVG_DC Quote
NewArpege Posted April 29 Posted April 29 10 minutes ago, KVG_DC said: Dude asked it to give reasoning for its scores too. https://docs.google.com/spreadsheets/d/1ICQD4pWp9hVhdFHpK_clxzVnCrkEM43p/edit?gid=1751986502#gid=1751986502 AI is jargon-ing so hard with so much confidence here... It doesn’t get it all right, but I’ve seen some absolutely insane and spot on comments come from it. For instance, I had it do Crowns show last year, and it said, “in part 2, the music sounds unstable and tense with trombone slides and yet the drill is moving in two linear blocks. There would be better GE if the drill behaved in a whiplash or has an earthquake scattering the boxes to better align to the sound of the trombones.” AI was right….and it’s early in its ability to do video analysis. It’ll only get better. Quote
KVG_DC Posted April 29 Posted April 29 Half wondering if its also found where to scrape existing judging tapes to pick that much lingo up and if it's actually generating these assessments or doing more of a copypasta of what it finds out there. 1 Quote
TheOneWhoKnows Posted April 30 Author Posted April 30 19 minutes ago, KVG_DC said: Half wondering if its also found where to scrape existing judging tapes to pick that much lingo up and if it's actually generating these assessments or doing more of a copypasta of what it finds out there. This. And providing more detail and repertoire of the shows prior to asking it to evaluate. Just as judges would be given. Quote
DieselDon Posted April 30 Posted April 30 Might not be too long before a corps will use AI to evaluate a show before the season starts and make recommendations. Then there's the issue of current CH and how they would accept AI. Quote
wolfgang Posted April 30 Posted April 30 1 hour ago, DieselDon said: Might not be too long before a corps will use AI to evaluate a show before the season starts and make recommendations. Then there's the issue of current CH and how they would accept AI. I cannot for the AI judge to hallucinate. "Your flutes are out of tune." 1 Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.