Loopy Pro: Create music, your way.
What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.
Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.
Download on the App StoreLoopy Pro is your all-in-one musical toolkit. Try it for free today.
Comments
@AudioGus
Is there a free way to sign up to CivitAI? I'm getting prompted for an API key for the Crystal Clear model, but there's no signup button just sign-in. Or do you have to go through the donation stuff?
I'm late to the game here (thanks @NeuM for pointing me to Suno.) I'm just as impressed as everyone with the quality of the results but just as I've started to explore it and refine my prompts I have been unable to use it as the free version is cut off due, as they write, "Due to heavy usage, generations are temporarily only enabled for subscribers." And now it always says that all day long. I haven't tried in the middle of the night. Wondering if I did subscribe if I'd run into similar road blocks.
As is, it seems more of a way to try out ideas with different genres etc. It's great to try out my own lyrics and if I can extract some vocals I might be using those with my own instrumental parts..... If I can ever get back on there again.
Hey @AudioGus and @kirmesteggno , you both seem pretty experienced with the graphic AIs and hoping you can clear up some stuff for me. So is Stable Diffusion a kind of open source environment where you must have it on your own computer to operate? It's not like Midjourney or Dall-E where I use a web platform access it. Whenever I've looked into it, the complexity is daunting to my feeble brain.
Aside from the usual prompts to create images, I've been interested in how I can upload my own images to Dall-E and let that churn out variations. Do you know if there are more ways to do that type of thing?
Also what about an AI video that can fill in-betweens of my own created images or video snippets. Any idea where something like that might be found?
Suno has fewer users after work hours (and even later at night), so if you are committed to using the free version, you will have to wait until there's a break in the paid service users. Based on the improved quality of recent results, I think I might subscribe for a month and see how it goes for me.
Regarding your other question, "Stable Diffusion" refers to a text-to-image model for generative art. It is used extensively by most of the services available today.
And from what I've seen, Sora is probably the most impressive text-to-video system I've seen demo clips of, but it's still not available to the general public. It has no release date yet.
Let's better use this thread for image AI: https://forum.audiob.us/discussion/49918/a-i-machine-learning-algorithms-to-generate-art/p22
I've replied to your question over there as good as I can. AudioGus is the man and knows much more about the topic.
Incidentally, Suno AI must be either completely overloaded with new users or they are having problems scaling because it is often unusable. I have been paying for this service and they just issued me a pile of credits because I was unable to use the service yesterday and again today I cannot get even the simplest thing accomplished... I dunno... I like Suno a lot, but I won't continue to use it if they can't get their service up and running for paying customers.
Hmmm, I binged my first months worth last week and was out but just checked and have 500 creds now. Noice. Seems to be working fine for me.
R.I.P Suno:
Let's hope the competition will have an effect on their terms too.
Like image generators these things are always playing leap frog with one another. The big jump will be an open source model you can run locally trained on everything under the sun with a much better text parser thingy... right now they just feel like text to speech remix generators. None of them can remotely follow prompts very closely from a sound perspective and are just coasting on the vocal and mainstream genre aspect. I want baroque bullfrog bango tango with hydraulic servo rhythm sections etc.
Haha, yeah. Btw, Stable Audio has audio to audio now!
I've attached a sample from a 90s sample CD and what the audio to audio made from it. The original sounds dirtier, I didn't use any lofi prompt for the generated one.
I had the slider that weights between prompt and source at 75%.
I also tried to sneak in a sampled loop from a 70s CTI record and got cought. They still subtract it from the 3 minutes you get on the free plan.
Oh my! OK, definitely worth investigating!
Looks like the new mystery one was https://www.udio.com/
Generations are currently down but may be good to sign up now as the beta may not let more folks in at some point yaddah blah blah
@AudioGus Where can you sign up?
I just hit sign in and then it let me use my gmail.
I'll pass until I can sign up properly with my preferred email. Udio is off to a bad start if they insist on those providers, kinda have my doubts if they're going to be more user friendly than Suno tbh.
It's all temporary throw away distraction for me at this point anyway. I want tools like control net for audio running locally before I get too excited.
The Grind Infinity track in the Hip Hop section sounds a bit like Group Home flow wise, but with a smoother voice. I could also hear some Metallica in one of the rock tracks 🤟
Guess I'm going to sign up through Discord 🥳
Udio is mind blowing, it's so on point style wise (thanks to NAS the digital savage 🤣). Gave it a prompt for a track about iPad musicians and added some tags to nail the 90s conscious Hip Hop vibe.. listen for yourself:
Well, this is pretty good. Udio is very capable of producing duets (something nearly impossible to get reliably in Suno, so far). This linked song sounds like First Aid Kit (in fact, I'm almost 100% sure that's what this song was trained on).
https://www.udio.com/songs/coixNX1gnJ1oWT8z2LQddk
(By the way, I won't be placing any generated song files here that I haven't created myself, only links to the work of others.)
And Suno just added a song exploration tool:
https://suno.com/explore/
Wow, adjusting the prompt and weighting using the remix feature is so amazing, actually feels really creative interacting with it. Sign of things to come.
How much time do your generations take? Mine anout 30 mins. What's weird is that some of the generations/variations I've downloaded immediately after their generation have disappeared from the app. Best is if you download the variations as quickly as possible.
Odd, mine take about 45-60 seconds to generate and are all in my library. Are you not logged in? Maybe they use different servers in different regions.
Yes, logged in, haven't used it today but yesterday and the day of release was around 30. I guess it's related to their (probably AWS) servers and the people accessing those. Maybe they've underestimated the demand in the EU.
Being a beta they probably have limited resources and maybe they cycle through different regions for priority and I have been lucky.
But yah, being able to make variations and change the prompt (and getting such quick feedback) feels so much like image to image in Stable Diffusion. This sort of thing will likely completely change how I make music soon enough just like how images have changed. Once there is something like control net with something like rhythm and melody canny/depth models (however that will manifest), will be amazing.
👍🏻 ai voices but what else is going on here?
It's great for sounds like vocals or Jazz soli that are a bit off limits for mortals, especially if you don't want to rely on sample packs and the stuff everybody is using.
Remixing AI songs by running them through stem separators and then adding to it could become a thing as well. Mixing engineers and people who like the traditional producer role where they direct the artists can use AIs instead and live out their creative side.
Same is going to happen with movies, where we'll se movies made by a single person and small teams. it will become very exciting with ML circuits built into CPUs and AIs into creative plugins and programs.
I've been playing around with this using Fadr, then dropping the stems in Samplitude and adding stuff with Maschine. Plus you get decent midi with Fadr which is cool. It definitely works but no doubt there will be nice clean hi-def multitrack generations soon enough. I am sure some people are using it now for real projects though, very useful, but yah once that bit of mush goes away and there are AI uprezzers etc that will be sweet.