Live data from Hacker News

Riffusion – Stable Diffusion fine-tuned to generate music

riffusion.com

121–130 of 481 posts

Re: Riffusion – Stable Diffusion fine-tuned to generate music

#121
post #33

Earlier this year, graphic designers, last month it was software engineers, and now musicians are also feeling the effects. Who else will AI make looking for a new job?

Politicians, bureaucracy. GPT-3, what policy should we apply to increase tax revenue by 5% given these constraints? GPT-3, please tell me some populist thing to say to win the next election, or how should I deflect these corruption charges.

Isn’t this the plot of Deus Ex?

Re: Riffusion – Stable Diffusion fine-tuned to generate music

#124

Other author here! This got a posted a little earlier than we intended so we didn't have our GPUs scaled up yet. Please hang on and try throughout the day! Meanwhile, please read our about page http://riffusion.com/about It’s all open source and the code lives at https://github.com/hmartiro/riffusion-app --> if you have a GPU you can run it yourself This has been our hobby project for the past few months. Seeing the…

is classical music harder? noticed you didn't have any classical music tracks. i wonder if it is because it is more structured?

Re: Riffusion – Stable Diffusion fine-tuned to generate music

#128

I’m curious about the limitations of using spectrograms and transient-heavy sounds like drums. It seems like you’d need very high resolution spectrograms to get a consistently snappy drum sound.

8GB is enough to do 1080p resolution. the UI i use for SD maxes out at 2048x2048. however, it takes a lot longer than 512x512 to generate: 1m40s versus 1.97s.

I'm guessing if one had access to one of those nvidia backplane rackmount devices one could generate 8k or larger resolution images.

Re: Riffusion – Stable Diffusion fine-tuned to generate music

#129
post #93

Other author here! This got a posted a little earlier than we intended so we didn't have our GPUs scaled up yet. Please hang on and try throughout the day! Meanwhile, please read our about page http://riffusion.com/about It’s all open source and the code lives at https://github.com/hmartiro/riffusion-app --> if you have a GPU you can run it yourself This has been our hobby project for the past few months. Seeing the…

All the AI music I’ve heard so far has a really unpleasant resonant quality to it. Why is that? Can it be removed?

I think this is because the generation is done in the frequency domain. Phase retrieval is based on heuristics and not perfect, so it leads to this "compressed audio" feel. I think it should be improvable

Re: Riffusion – Stable Diffusion fine-tuned to generate music

#130
post #127

deleted

I think you'll find plenty of people who find that DAWs and music theory help them better find self-expression and celebrate life through their music. Any tool or framework that opens up new modes of achieving that self-expression should be celebrated, not shunned because it isn't as "pure" as more time and labor intensive methods. Would you rather someone be forced to dedicate a significant amount of time to studying music and art creation just to be able to find that self-expression?
Post reply on HN