3 ms·
I tried generating music on stableaudio.com and, yes, it's bad. However, given the blistering pace of developing in these models, I would not be surprised if th
by laborcontract 3y ago
I tried generating music on stableaudio.com and, yes, it's bad. However, given the blistering pace of developing in these models, I would not be surprised if these sound incredible in a year or two.
- berkes 3y agoEveryone every time seems to assume a linear (or exponential) curve upwards. But what is the proof for that? I consider it far more likely that we had a breakthrough and now rushing towards the next plateau. Maybe are nearing that. Like in the curve of a PID controller. It's how most or many human improvements go.
- leodriesch 3y agoI'd say most are thinking of Midjourneys success in image generation when talking about this kind of progress.
- berkes 3y agoI'm too. But I still see no evidence that this keeps improving and not plateauing at some (current?) level.
- spacebanana7 3y agoThe plateau we're heading for is getting professional human level output from these models with logarithmic progress. I suspect this is because the underlying production factors like compute, data & model design are steadily improving whilst humans have diminishing sensitivity to output quality. In the game of AI generated photorealistic images or history essays there's not much improvement left to make. Most humans are already convinced by the output of these things.
- laborcontract 3y agoI think the proof is seeing how good diffusion models have gotten for making images. They're not perfect but they're leaps and bounds over what we had just a year and a half ago. Many of these problems seem to have been unexploited simply on basis of nobody throwing enough gpu clusters at it yet.