Re:ESPN AI-generated recap of retiring player's final match ...
by regretaverse on Monday September 09, 2024 @12:54PM
(#41490526)
OpenAI transcription & LLM is likely more costly than what they're willing to splurge on this. Care to link soccer.txt, so I can try with Llama 3 / 3.1?
[ Parent
| on HN ]
Re:ESPN AI-generated recap of retiring player's final match ...
by jetrink on Monday September 09, 2024 @01:17PM
(#41490873)
Here you go: https://pastebin.com/QXzg95AV
One thing to try is only using the post-match commentary, which starts after the line containing 'final whistle.' When I did this, the result was more factually accurate, while still focusing on Morgan.
https://chatgpt.com/share/9c122702-b46f-4e5e-bd05-e063a74126...
Re:ESPN AI-generated recap of retiring player's final match ...
by whimsicalism on Monday September 09, 2024 @01:05PM
(#41490700)
too costly? it’s probably a few cents total
Re:ESPN AI-generated recap of retiring player's final match ...
by brewdad on Monday September 09, 2024 @01:31PM
(#41491054)
And it got the most basic detail, the score of the game, wrong. Even at the cost of only a few cents it's worthless.
Re:ESPN AI-generated recap of retiring player's final match ...
by whimsicalism on Monday September 09, 2024 @02:24PM
(#41491720)
seems like this is a solvable problem with just a bit more engineering effort, but yeah... totally worthless
Re:ESPN AI-generated recap of retiring player's final match ...
by acchow on Monday September 09, 2024 @03:46PM
(#41492727)
You should just need a better prompt. I think everyone would benefit from using a standardized prompt which asks the model to think through its work between `<thought>` tags before writing its response, and also reflecting on the response between `<reflection>` tags, and then outputting the final response afterwards