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.
AI generated music VS Music created using randomised patches, midi & drum sequences?
What’s the difference?
Comments
Current AI output is trained on huge imports of existing art… hmm… so is most pop art.
True innovation [edited here] ignores existing rules or patterns: if anything it’s violations of rules and non-traditional
forms.
Cubism, abstraction, pointillism…
AI generated music would typically be created according to rules or algorithms or patterns for generating music that sounds good and makes sense. There might be element of randomness, e.g., in choice of key -- or in choice of specific generating algorithm, but the music itself will be based on rules known to generate listenable music. A randomized sequence, not so much. If truly randomized, the sequence may or may not sound good or make musical sense, probably not.
Piano Motifs is an example of a kind of AI music-generating software. Ask its creator if he intends for it to be giving you "random" sequences and I'm sure he'll tall you flatly, "No". It follows common rules to generate musical pieces based on basic parameters set by the user. It take a good amount of musical knowledge to create a reliable and useful tool like Piano Motifs. (RE: some randomness -- I can't remember -- there may be some control(s) in Piano Motifs that controls level of randomness, and where you're likely to get less and less usable results as you increase it, though small amount might give you something unexpected, but still sounding good.)
We’re really in trouble when AI can truly detect if its output is derivative and it re-calculates another solution seeking something that doesn’t match the model. Of course, like the first time you hear a great artist you’ll be inspired to clone that sound.
I’m just wondering. The use of AI tends to be frowned upon here on the forum, at least that’s the sense I get - but the randomising aspect seems to get an easy run in comparison.
Yeah, it’s a fair point. Quite honestly, I feel many negative comments about AI here on the forum have been poorly informed and from people who haven’t spent much time trying to work with the tools available. I mean that last comment in a general way. There has, after all, been very little truly AI-produced music posted here. Most discussion has been about text or image generation.
The few AI music apps I checked a few months ago genuinely didn’t seem impressive. That field seems way, way behind image and text. Audio quality also tended to be extremely low. If there isn’t much good stuff around, there will be soon, though. And I will be very happy to use bits and pieces of AI generated stuff in my musical noodlings, I have no attachment whatsoever to the idea that one should make everything oneself from scratch. Still worried about AI on a societal level, but I think most creative people, at least in the digital domain, will embrace it in time, in their own creative endeavours. Certainly those who enjoy working with generative midi have no reason to prefer simple algorithms over AI, that just doesn’t make sense to me.
I like it for making samples, mangling it with effects and mashing it up and jamming it out with my other ‘real’ 😀 iPad instruments. Or even generating quick rhythm backing tracks etc. I still feel quite creative 😛
No difference. I’ve never understood people using generative midi sequences nd stuff. Nothing against it but I like writing music and that is by definition, objectively, not writing your own music. But different people. I love ai - a lot more in fact it’s way more interesting as it takes it further. An it’s a big glorious f u to the state of music today But the best yiu can say using generated main melodies or sequences is yiu helped arrange sequences generated by an algorithm.
It’s about how much involvement you want in writing a song. How much of the song is coming to you.
TBH it says something about the state of music today this needs saying
If you use it to generate a drumbeat or a bass line of course it doesn’t mean you didn’t write the song unless that song is drum n bass lol it’s the equivalent of getting a session musician in but even then i don’t like doing that as it really misses what i feel im hearing in my head and what the song should be at its best, its just making it ‘work’.
I think it’s better for the types of music that are more about production than composition. Which is basically everything now
I don’t know how anyone can frown upon using ai then in the same breath be ok to use any kind of generated midi sequence (where that’s the case) which is commonplace in production and electronic music. The hypocrisy is laughable. The reasons are so transparently that it touches a nerve in terms of the burlesque of their own musicality. They should just get over not being songwriters.
I think suno was the ai i tried. When I say tried, I mean, I made it write loads of incredibly offensive, probably prosecutable far beyond politically incorrect songs, with a friend, to amuse ourselves. It was great. I agree mostly about music, but every few songs there would be a particular line or passage that worked really well. The issue is that it doesn’t yet know that. So someone could easily use it to create the soul of a song and then write the rest of it work with it even back and forth. For me, I can do all that better and far more quickly myself but for others that’s not the case (I get that might sound arrogant but it’s just reality).
I find it quite disturbing lol. I don’t want to do it myself - I’d rather shoot myself. However, I don’t find it wrong. Im well up for ai obliterating us, it’s long overdue, and I lol greatly at its trouncing the artifice of most industries and peopel that call themselves creative. I have to admit though it’s well disturbing when it treads on your turf.
But, my principles and my ego are well matched. So I view it as a worthy future addition to songwriting.
For me, as someone who writes, thinks, and writes music and really have no sense of worth beyond that, on the one hand its kind of a shock (half really funny half devastating lol) and on the other it doesn’t bother me. Because I already really like what I do and it’s my purpose in life. So if someone else, whether it’s ai in the future or (less likely
) another human does something cool I see it as someone else doing other cool stuff. It’s not preventing me doing mine. It’s good, it’s a reminder not to get lazy to keep pushing myself and not to let whatever talent I have lie in the corner gathering rust just because it’ll still do a job. That’s nit good enough. It’s nit about being better than anyone else it’s being the best you can be, it’s a personal journey that has nothing to do with anyone else. If you are at the top of your field and noone comes close, you still need every day to come at it with the same love nd the same sense of discovery and improvement and wonder as you did when you started. Anything else is inauthentic and that’s nit good enough
Anyway before all that self-aimed bullshit I was saying that suno ai, did among all of its wayward efforts create one amazing song and it PISSES me off because I fed it my own lyrics. My friend loves it -_- I played him the original song, that I wrote and he wound me up no end by going ion about the other one.
Of course, he did later say begrudgingly that my one was its equal `(…) nd ask for a version. I want to be clear here… he literally thinks the ai one is one of his favourite songs of all time, so this is quite an accolade. I genuinely don’t think even he knows how much of this is true and how much is to piss me off
EDIT: actually put it this way. If you’re doing something that’s replaceable by ai, then you shouldn’t have been doing it. But that isn’t because you’re bad or good at it, that’s because you placed value on it being sellable or successful rather than doing it because you love it because that isn’t replaceable by ai. Because What does ai love? Nir will it ever be replaceable by ai because at the point ai loves something it is no longer ai. And love for things is a scale toppled onto its side where there is no comparative worth just the colours of value
Sorry for the multiple posts but on the subject of AI.. I feel like something will happen soon because it seems to me at least that it’s plateaued. I’m sure there is a lot of development but it’s like video games. Development and innovation was wild and rampant in the days of amigas and things. Then as this new technology sort of settled into formulas and use cases, development started to be shaped by those structures rather than shaping the structures. Games are almost all generic now by nature. In those days games routinely felt in between genres and exciting, same with all industries films books
And I feel it’s an illusion. Genre doesn’t exist it’s a perception. But we then start shaping things to it. It’s an impression left by fluid things not their container.
So like uhhh yeah ai… it’s happened somewhat. I think the rampant development resulted in some kind of ecosystem of usages which has kind of dissipated development somewhat it’s no longer shouting it’s talking intelligibly but quietly and less excitedly about more specific things,
So I think, if I can invoke some kind of hopeful prophet, that this is a situation in which a leap might happen. Just like the one from which ai sprung,
My problem with it is the applications aren’t ambitious enough. I have one in particular in mind which is possible and would … ahhh someone who can develop some amazing ai stuff come work with me
No doubt some developers are currently working on some kind of midi AI where a prompt describes something and it churns out midi. Also, I’m fairly certain that the major DAWs will all be adding AI tools directly into their software just as Adobe is doing now with Photoshop, etc.
I feel that actually playing an instrument using your physical body e.g. hands, breath, etc. is the most pleasurable and satisfying way to make music.
But I also really enjoy using digital tools, arps, sequencers, etc. to generate ideas and pick out useable ideas to build musical compositions.
And I’m not conflicted about that at all. Composers, musicians and artists have always used randomness and chance to play with ideas. Play and chance are always intertwined.
Leonardo said he could get ideas from splotches and stains he might see on a wall. Composers have used “games” like throwing dice and using letters mapped to notes to make musical phrases.
If you just turn on a switch and let random notes or beats play on and on. What’s the fun of that? It’s boring and meaningless.
As long as I’m picking and choosing among the random stuff that’s generated, keeping some and discarding the rest, then I’m still expressing something about myself.
Just complexity of the output. You generate enough variations or roll the dice until you have something that works for you.
👍
Yeah, what’s the difference between using AI generated loops or samples, as opposed to a sample library? Are people that use samples other people made not musicians? Did anyone think sample libraries would kill it for musicians because now bedroom musicians could replace them?
I've no opposition to either random or AI. People can use it all they like and I think no less of them. But to me both feel empty. Music for me is the joy of creating it out of my own self. I end up feeling just a little cheated to the extent that I use tools that do that for me.
That said, I do use degrees of both. I think we all do. For instance, I sometimes turn to the Lumbeats apps for help with drum parts because I lack the time and skill to do well myself. That is nowhere near as satisfying as doing it on my own, but I have to admit the results are (usually) better.
I've made songs using Riffler because I think some of the riffs it generates are cool as heck. But the results leave me feeling empty. I can enjoy an evening fleshing it out, but the guitarist in me is bored and feeling left out.
Knocking up a generative patch in miRack is satisfying, but in that case I'm deeply involved in the process and the tweaking of it. I'd have no interest in using someone else's patches.
I have absolutely zero interest in totally push-button results. It comes down to the results being less important to me than the process. If results were the main goal, then I guess I might feel differently.
Generative stuff, on the other hand, was something I used to think as kinda like cheating. Took me a long learning about the beauty of modular to come to appreciate it (and nowadays is what I have most fun with)
As for the OP question: to me AI is more focused on projecting something based on patterns from the past. In that sense it's very much like the way we create music. What we create is generally a synthesis of our own influences. I think there is such thing as the muse, though, and AI is lacking that.
Randomization is just ... random ... there's no pattern synthesis behind it. It can make some stuff that's pleasing to listen to. I get bored with generative music pretty quickly though unless I made it.
@wim I can only stand my own generative music too (and no one else does)
Sometimes the process is the art. In fact, always.
Random can be useful. I like using the % chance of playing steps in step sequencers, random order arpeggios, sample and hold LFOs, etc. Random can usually be guided or constrained in some way to get good or useful results. Besides, isn’t random in computers just an illusion?
It all depends on the boundaries/model defined by the developers using training data and prompts as the parameters. A good randomizer "knows" the program it is embedded in, it has a range that was tested by the devs within the contexts of music genres and will often output useable results, so does AI.
Midjourney produces more artistically pleasing results than other image generators because the devs are more competent in Art and have made better decisions regarding training data and parameters.
The training really is the creative side of it, it will only be as good/artistic as their teachers taste.
Could one draw a distinction between “random”, generative, and AI? I think so.
In at least how I’d define things “random” is truly that random in what it produces like a cat on a piano or recording one’s surroundings. Generative try’s to impose rules on random by scale rhythm both and perhaps many other rules some of which could be rather arbitrary. AI gets trained on past examples made by its progenitors, us, and uses our vocabulary and past success to base its compositions on - one might say rather well but formulaic.
No matter which approach is used it is us that either pre (in editing composition etc) or post (in appreciation popularity etc) imposes order and - let’s say in a rather broad way - meaning and appeal to music
I’m neither for or against any of them. If you want a primer and some thoughts on what we’re going through to some degree with AI look no further to the book “Art and the age of mechanical reproduction”
I think people use the term "random" very loosely, and if a tool does indeed generate pleasing music it is never totally random, or even close. Constraints need to be added to any tool that's generating music, and to the extent you add constraints you diminish randomness. To some extent people can use (and get satisfaction from using) their musical knowledge and expertise to get good results out of "random" generation by specifying intelligent constraints. (And sometimes the intelligent constraints are built into a tool that claims to "randomly" generate music; in this case it can give good results even if the user is a chimpanzee, knowing nothing about music.)
To illustrate something slightly closer to true randomness than what you get from any tool that is remotely useful: generate a midi sequence, in which the tones will be randomly a midi note from 0 to 127, each having a random velocity from 0 to 127, each note with duration randomly between 0.01 second to 100 seconds, and polyphony will be randomly 1 to 127. Attach the sequence it generates to a synth producing the most beautiful sound you can imagine. How long will it take to generate a pleasing musical song? I don't know. How long would it take 100 chimps pounding away at typewriters to produce Hamlet? I expect these lengths of time are on the same order of magnitude.
Back too my original point: How/why can people get usable results from tools that "randomly" generate music? Only because and to the extent that intelligent constraints are placed on the "random" generation.
The intent of the user is the big difference. AI and Random generative apps both tend to have rules. AI has more complicated rules and likely a dataset to… coughs take inspiration from.
The term “AI” is used quite loosely nowadays. A computer program making pop music might not pass the Turing test, like many of today’s pop stars (ooh shots fired! I’m only joking!)
That's not AI. That's generative music.