this post was submitted on 14 Apr 2024
60 points (91.7% liked)

The Lyrics Game

425 readers
1 users here now

Anybody can post a challenge.

The rules of the game are simple, take any song you like from any genre you want, put the lyrics through any AI [you may weight lyrics as the AI allows] that you want and post the resulting image here under the post title Name That Song [GENRE]. After 48-72 Hours or if the answer is guessed sooner, edit the post's title to [Solved][Genre] and put the song title and artist (and optionally any highlighted lyrics) in a spoiler tag in the body of the post.

Other than that, enjoy.

Here is a list to a few AI image generators you can use.

founded 6 months ago
MODERATORS
 

First Post (but have loved playing along)

you are viewing a single comment's thread
view the rest of the comments
[–] Voyajer 4 points 5 months ago (6 children)

Oh no! 5 seconds of GPU time on consumer grade hardware!

[–] [email protected] -5 points 5 months ago* (last edited 5 months ago) (5 children)

It’s not nearly as small as you think it is.

Generating 1,000 images with a powerful AI model, such as Stable Diffusion XL, is responsible for roughly as much carbon dioxide as driving the equivalent of 4.1 miles in an average gasoline-powered car.

https://www.technologyreview.com/2023/12/01/1084189/making-an-image-with-generative-ai-uses-as-much-energy-as-charging-your-phone/

Even better when you take into account the scale at that these run at:

https://www.washingtonpost.com/business/2024/03/07/ai-data-centers-power/

[–] Voyajer 3 points 5 months ago* (last edited 5 months ago) (3 children)

I know exactly how much goes into it, 5 seconds of GPU time, on my own computer. That's why I said it. How many phone charges do you think it would take to fully create a digital drawing on a laptop? It's not going to be much different IME.

[–] [email protected] -4 points 5 months ago (2 children)

You aren’t taking into account the resources required to train the model. You clearly have very little idea how much goes into it other than running software someone else wrote.

[–] ForgottenUsername 2 points 5 months ago

We aren't training models here, this isn't the training-ai-with-massive-datacentres@lemmy

[–] Voyajer 2 points 5 months ago

Of course I've taken into account model training costs, was that supposed to be your gotcha? You don't actually think the energy cost amortization from training still accounts for the bulk of energy expended per image generated do you?

load more comments (1 replies)
load more comments (1 replies)