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 starts to be serious
This discussion has been closed.
Comments
Irrelevant.
Creativity does not equal emotion.
If you think that line of argument is valid then you and I see the concept of human creativity too differently to really have much of a discussion about it.
What ?? I’m not sure we understand each other :-))
You said:
“ Until AI develops true emotions (as opposed to algorithmically simulated ones) ….”
So I am asking how you define true emotions as opposed to alhorithmically simulated when it comes to brain vs. LLM.
You're right. I did misunderstand you.
But I'm not interested in engaging. I think there's too much of a disconnect in how we view emotion / creativity. I believe there's a spiritual component to both that you seem to discount.
I was clear that it can be simulated effectively but not replaced. Simulated being the key word there.
I don't think emotion is the barrier. My perception is that the biggest barrier faced by generative AI (LLM) is the fact that it is language based. Even when you input audio or visual media, it processes it into language, and its output is also language, processed into audio or imagery. Now that's obviously a huge difference from a human brain, which does not convert all its input into language, and does not use language for the majority of its output. If I listen to a song played on a guitar, and then pick up my guitar and try to play it back, language isn't involved at all.
Why does that matter? Well, language is actually a bottleneck in many respects. It is an extremely narrow way of processing the world. Imagine you live in a tiny windowless cell, in the dark, and the only communication you have with the world is by having someone describe it to you. So they describe to you a bunch of paintings, and ask you to describe a painting back. Okay, you have all this information, so you can describe it. But when they ask you to describe a totally new painting, you have no idea what that means. Because you don't have any visual information, you've never seen a painting. An LLM has never seen anything, it has never heard anything or felt anything or tasted anything. It has never felt gravity or magnetism. It has never felt heat or cold. All it has is a symbolic language. And there's no simple way of getting around that, because generative AI is language. You might as well say that we need a building made out of water, or a car made out of butter.
So I don't think LLMs are going to lead to super intelligent AI in the near future. I think they will lead to other models and architectures, which may be amazing, or not. But language alone is not sufficient for agentic intelligence. And although AI is clearly getting better at mimicry and synthesis, it will always be one step behind creatively. And in many respects it is still far, far behind. Like, no kidding it can create a pop song, a freaking five-year-old can make a pop song. Ask it to make Moonlight Sonata.
Great post @timfromtheborder. 😎
Surely LLM based systems aren't the only form of AI being developed however, are they?
As they say, right now is the worst it will ever be and it will only continue to get better. People will have no problem with generated things because they’ll be indistinguishable from real.
That may be true by and large, but I also believe there is a spiritual component to human connection. I think people will still crave that and recognize it at some level in creative works.
I think the word "spirituality" rather than "emotion" would have better expressed my thought.
Anyone here who says "This is all just AI slop" clearly feels emotionally threatened by the power of AI on a subconscious "lizard brain" level, because this is just objectively false. The quality of most AI generated music, even just simple prompts to Suno etc., is objectively on par with or exceeds most commercial offerings in the charts.
And the technical quality -- the highs in this video, as mentioned in the thread -- is absolutely fixable and might actually be a compression / transcoding artifact. All songs I've had Suno create for me had better audio quality than this.
Bullshit. I just don't like the music. I would feel the same if it was by a human. I say it's AI slop ... because to me it is slop, generated by AI.
The music anyway. The video does bother me at a "lizard brain" level, I think because it's something trying to appear human, but my inner sense knows it isn't. Kind of like motion sickness.
I do believe I would suspect the music was AI even if I didn't know ahead of time, but could be wrong on that point. Too late to tell now.
Can't argue there.
This is a fundamental misunderstanding of how these systems actually work.
LLMs don't "process everything into language" -- that's not how the architecture functions. They work with something called "embeddings". Numerical vectors where the same token gets completely different representations depending on context. "Key" in "musical key" vs "door key" becomes entirely different vectors after processing. The model never "sees" words as words -- just learned representations that capture semantic relationships. i.e. "meaning", if you will.
Also, modern AI isn't limited to "language" at all. Multimodal models process images, audio, and other data types directly as their own embeddings, i.e. not converted to text descriptions first. Vision transformers work with image patches, audio models with spectrograms, etc. LLM just happens to have "Language" in the name because that's where it first got traction. It's a historical accident.
Your "windowless cell" analogy assumes these systems only have text, but multimodal models are literally trained on images, audio, video, i.e. directly on sensory data, not just descriptions of it.
The grounding problem (whether learned patterns = understanding) is philosophical. But claiming the architecture is "language-based" and can't work with other modalities is wrong.
Yes, OK, it's totally fine not to like the music 😂 I was basically referring to those folks who just knee-jerk "AI Slop" reflexively every time they know something has been created by an AI. i.e., "ideologists", if you will.
Yes, I understand what you mean and I actually agree after watching the video. It's in Uncanny Valley. But this will improve a lot in no time.
And let's face it -- it is still, on a technical level, utterly insane. This would've been a $100,000 video production just 1 year ago, now it's just "slopped" out of an AI's... ass? 😂
No, but generative AI, what we're discussing with Suno, ChatGPT, Unstable Diffusion et al, is LLM.
This is equally an appeal to emotion. First off, there's nothing objective about musical quality. There is no measure of a good song or bad song. From my perspective, there is good music that I like, and good music I don't like. There's also bad music that I like, and bad music that I don't like. And that is different for every single person.
Second, I don't agree that AI generated music is on par with what's on the charts. I'm not a fan of pop music, but I get exposed to it enough, and I recognize that there is a lot of talent and creativity involved. How anyone can say that this Suno generated stuff is on par with Chappel Roan or Dua Lipa is beyond me. I just don't understand it. Again, that's your opinion, but I think it's totally wrong. I don't care if AI music is good or not, it doesn't affect me in the slightest, but every bit of it I've heard has just been bad music. The video in this thread is bad music and anyone who thinks it's good has awful taste in music, sorry.
As to your subsequent post, I understand that LLMs aren't using "language" like we use English. But they are still rooted in symbolic language processing and are not trained "directly" on sensory data in the sense that a human brain processes sensory data. I understand how the architecture functions, perhaps I should say "symbol based" rather than "language based."
That is an important point. I wouldn't be surprised to see that change. It's probably already being done or at least being worked on.
That Rick Beato video a few posts up doesn't get to making any points about music until about 10 minutes in. The real point starts at around 14 minutes.
Skip to 14:45 for a real nugget.
what a emotional and intense discussion I intriggered here 😂
Good, it feels this forum is still full of humans instead of spiritless LLMs 😂
Perfect time for your comment, cause…
https://m.economictimes.com/news/new-updates/internet-is-dead-reddit-cofounder-alexis-ohanian-says-bots-now-posting-content-generate-fake-engagement-and-even-simulate-conversations/articleshow/124596816.cms
It’s time to close the doors to Loopypro.com @Michaei, unless while peering through the peep hole you notice them wearing a pleather Members Only jacket.
Always zoom out/take a step back and see the whole picture.
It's very healthy, I say...
To quote the laconic Adam Curtis on this matter.
"In the past, music was how people made sense of their lives, a way of saying this is who we are, this is what it feels like to be alive now.
But somewhere along the line, that stopped mattering. Music became background, algorithmic wallpaper for a distracted age. And just as we lost the ability to imagine anything new, the machines arrived to do it for us.
We don’t need more derivative music made by machines; humans are already doing that in copious batches. Entire genres now exist as simulacra of themselves - infinite remakes of a mood, a nostalgia wink, a preset. The tragedy isn’t that AI imitates us; it’s that we’ve spent years preparing the ground, fueling it with our own creative exhaustion.
AI music isn’t about creativity. It’s about control. It turns emotion into data, and data into profit. The people who win are the ones who already own everything else - the platforms, the feeds, the attention.
And yet, we are seduced by the mirage. By the promise of infinite songs, personalized emotion, music that knows us better than we know ourselves. It flatters our loneliness, disguises our boredom, and feeds us a simulation of feeling. The machine doesn’t need to understand us - it only needs us to keep listening.
The capitalist industrial dream of endless music or endless everything wasn’t freedom.
We mistook abundance for emancipation."
plus my note (2 (n)eurocents?!:) on the very spot-on timfromtheborder point:
Wittgenstein once wrote that
"the limits of my language mean the limits of my world."
He understood that language could map reality, but only up to a point -
beyond that lay experience, emotion, the ineffable, all the things that resist being named.
( and all of the above named change every nanosecond!!!).
Some say that sound is another kind of language. But it isn’t.
Sound is a medium, I feel - way older, way deeper, untamed by syntax, way more granular.
That’s why *language-based models will never understand what music does,
it will just frankenstein it - too many intricate unseen or barely seen vector webs to simulate.
Because sound (and music) doesn’t just describe the world.
They create it.
*language-based models are after all just brains in a jar (vat).
There’s an irony: in using LLMs, humans risk becoming brains in a vat themselves,
mistaking language for reality and simulation for experience.
But unlike LLMs, we still have bodies, senses, and stakes in the world.
We can ground meaning, they cannot.
Their “judgement” is only an echo of statistical patterns, not understanding.
https://iep.utm.edu/brain-in-a-vat-argument/#:~:text=The Brain in a Vat thought-experiment is most commonly,experiences of the outside world.
I wonder what the AI generated songs would be like on the topic of the end of civilisation brought about by the use of AI. Wouldn’t that be something?
Hear hear
AI already surpasses humans on a variety of language and visual tasks and shows no signs of slowing down.
Already we have AI systems that the are equivalent in intelligence to someone with a PHD. And not just in one subject but every subject and they are learning all the time and billions of dollars are being poured in by massive companies across the planet to win the AI intelligence race. Sorry to say this but folks who think AI is slowing down have no idea of the global effort to be first.
I’m a self confessed AI doomer. I think in the short term (next 10 to 20 years) AI will have a devastating massive impact to humans related to jobs. It will force us to ask what it is to be a human when we are no longer the smartest intelligence on the planet.
cyberheater:
AI doom is a scabrous western luxury panic.
The real apocalypse is already here - climate collapse, inequality, endless war.
Fantasizing about machine takeover just helps us forget
the systems already destroying us.
I genuinely wonder how an intelligent species has allowed this to happen. Maybe our hyper intelligent AI overlords will do a better job.
Our intelligence has not evolved to perform well in societies of this scale and with the massive power that we've developed. It makes us totally incomparable to any previous apex predator. Highly recommend watching some Daniel Schmachtenberger videos on existential risk.
I get you with climate collapse. But surely the levels of inequality and the extent of war today are less than they have been historically, over last 10,000 years, no? I'm thinking of Steven Pinker's 2011 book, "The Better Angels of Our Nature: Why Violence Has Declined", which makes a decent case for that.
I think Yuval Harari identifies (1) climate collapse, (2) possibility of nuclear war, and (3) technological disruption from advancements in AI and bioengineering (many of these technological disruptions relate to inequality) as the three biggest worries. Seems like a good list to me.
Well said.
hes:
if you want, I can debunk Pinker's book to bits.
but don't have much time.
I suggest you read some David Graeber (and Nassim Taleb, John Gray etc)
and discover yourself why Pinker book is wrong at any step.
And dangerous. And how west lives in that "Things are getting better, thanks to us.” feel-good & propagandistic lie.