Audiobus: Use your music apps together.
What is Audiobus? — Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.
Download on the App StoreAudiobus is the app that makes the rest of your setup better.
Comments
The most alarmist ones tend to be those who haven't bothered to research what diffusion rendering actually is and have never spent much time using it themselves. They just hear about the basic idea of typing what you want to see (like a google image search), look at the results, then draw massive conclusions about what it is doing, particularly that it is just cutting and pasting and rearranging bits of existing images. If they only knew the far more alarming truth...
Wow! Disturbing!
In the case of generative art, the connection to copyrighted sources should be easier to prove than in real life.
That is, either the algorithm trained on an artist's images or it didn't. If it did, then the generated art is a derivative work, is it not?
I expect and hope that copyright law will be enforced in generative art in order to compensate the copyright holders for these derivative works.
This will result in some accountability for the derivative work, and some small compensation for the original artists.
https://www.legalzoom.com/articles/what-are-derivative-works-under-copyright-law
Maybe this will depend on how 'training' on images for a machine differs from a human 'training' on images.
What happens when a derivative work is completely visually distinct from the dataset sources though? Most of the things that come out of a given diffusion rendering bare no direct resemblance to any individual pieces in the dataset. In my mind current laws regarding copyright and trademark infringement are that a given work be visually obvious as such. So yes, if you use a diffusion renderer to recreate an established Pokémon character and then you sell it, sure that is trademark infringement, no different than if you made it by hand yourself, because it looks like an existing IP. If however you create images with diffusion rendering that do not look like an individual work in the dataset (the vast majority of renderings), where is the visually obvious infringement then?
Perhaps laws will come in where businesses need to disclose their datasets if they profit from machine learning of copyright works but I am sceptical that anything on the books support this now and that it would happen any time soon. For individuals generating their own diffusion renderings with their own offline dataset (which will be extremely commonplace and dominant very soon) there is pretty much no ability to govern this. So I feel they may start to inch forward on new laws but the fact is most people will be able to entirely generate these things independently. The current Stable Diffusion model is an open source free 4gb download and renders on local hardware.
Thanks so much for the info @auxmux
For now on I shall take pictures of something and tag them as being a picture of something else
how did you guys managed to get into Dall-E? one week waiting list... took it so long with you also?
https://replicate.com/methexis-inc/img2prompt
It can deconstruct the content of an image without tags
But if enough photogs and artists fuck with it, such generative prompt driven systems won’t have a leg to stand on, at the 51% point they’ll have to believe us, and we’ll have won
The datasets have billions of properly tagged images now to test any new images off of that it encounters. The battle was lost long before it started. You'd need one of these...
Just getting over covid and been playing with MJ….. fun times.
I find all of these images disturbing and hellish. Even the more benign ones are somehow disturbing. I have no ideological beef with AI; this kind of art is just engendering a very weird vibration for me. Even when it’s good.
@Wrlds2ndBstGeoshredr not that these images will help you with your vibratory interpretation of these images, but i will now give you what AI depicts with:
Visions of the Somnobulist (for i too feel that these images lay somewhere between that state of wakefulness and dreams):
These are great. Some of this AI art is just absolutely mad trippy. What software was used to make these?
@Gavinski yeah it is truly wild. These are done with MidJourney. I find the most success with it for my noodling.
Thnx yeah I joined their discord yesterday, seems to be regarded as one of the best, cheers
Welcome to the Rabbit Hole of infinite variety 😵💫😉
Some recent experiments.
I forgot how long I waited, but at least a week for me certainly, maybe longer? I don't have IG, Twitter, Linkedin. Maybe that affected their selection algo.
Yeah I am just a human too - no influencer and stuff - but I got my invite this night. WOW, this is amazing!
Right now am working through the the biggies (Dall-e, MJ and NightCafé). I find nightlife has with stage diffusion a winner too, but you have to go to the advanced options and turn that prompt %% to 100%. Was not happy with the outcomes but when I started to turn this slider up. wow, great results!
I wish MJ would have a 5 free shots a day policy too
I simply signed up as a regular person not an artist. I didn’t add any social or anything either. About a week for me with Dall-e
I’m absolutely with you, there is something that is not “good”, it bring something that not belong to this world and I deeply feel shouldn’t be here
Yes and no.
). Many many people somehow longing for transcendental material, demons, cyborgs, fantasy beings and stuff. Let me post some of mine - are these carrying the hellish dark flavor too?
I get you both, but it’s „garbage in garbage out“ (it should be clear I am not talking about garbage
I wouldn’t say hellish myself, more disquieting and ominous, sinister in some ways perhaps. It’s interesting that they seem to share this quality no matter the subject. I wonder if the matter of ‘soul’ could be invoked in the discussion of them. (ghost in the machine style rather than Old Testament religions). Is a machine making images going to be able to create anything invoking joy for instance?
Maybe there’s a bias, in knowing who the creators are?
Always the case, but being as objective as possible I still see that trait. I think there’s an interesting discussion to be had about what humanity is arising from this technology for sure though.
Absolutely, not just limited to it’s use in the art’s though.
https://qz.com/the-next-step-for-dall-e-and-midjourney-is-animation-1849462801/amp/
Just storing this here, I could do with a research grant for this 😁