LOCK UP YOUR MUSIC THE AI IS COMING! Which allows me the opportunity to ramble on about stuff that’s been on my mind for quite some time.
Great video. I also love seeing workflows and am normally more invested in the process that people use rather than just the output itself, so if this becomes a little more common in the midst of AI, I’ll be extremely happy.
Whether it’s good or not is debatable for sure, but I’ve been having a blast giving out ‘secret sauces’ for the past few years and it’s interesting that this might actually become a little more important in the future, even if it’s just for the few who care about the creative process.
Since this is the renosie forum, I think it’s safe to say this without offending anyone here, but when the masses all flocked to using presets and the same boring loops as each other, I’m pretty sure the only people who were tricked by it / actually felt something unique in hearing the same shit regurgitated over and over again were just deluding themselves anyway. I suspect something similar might come with AI.
I say, weed out the people who don’t actually want to make music and let’s get back to being creative
Reminds me of the “Epic Mash Leg” series of joke mixes featuring the “top 100” Beatport bangers…
Many others have rinsed out variations of this meme in slightly different ways over the years, so you can find plenty of other examples on SoundCloud and YouTube.
Lazy Wanker Music for Lazy Wankers
You make some good points in your video. Though I’m not sure the “people will assume your work is in in part, or entirely is made by AI” is really anything new. People already assume working with a computer means there is a “make this awesome”-button you’ve pressed. And to some extend that’s true for a lot of people making music these days.The sound quality of the sample packs available today is top notch. Not to mention the presets in synths. It’s not hard at all to make some generic yet professional sounding track in a short amount of time.
To me streaming the making of a track feels like content for content’s sake. A highlighted, condensed version of what you did is great.
Recently I tried making a video of me recording my synths. People are always interested in seeing hardware. Just pop that video on screen when that particular synth is playing in the final track. They’re happy seeing the synth in action and I’m happy as I’m getting them to listen to my music. But for me the filming got in the way of creating music. So I abandoned it and use pictures instead. But more power to the people who can do that easily.
Anyway, allow me to plug my video on the influence of AI on visual arts examined through the lens of Ghost In The Shell:
Thanks for the video!
I imagine one day AI will generate songs in Renoise too - perhaps even real time.
What will humans do? Idk - jam with acoustic instruments on street corners and parks.
Really? As long as people suck at composing and/or mixing, it doesn’t matter how good samples or presets are. You need to know how to use them properly, otherwise your production sucks anyway. And I’m pretty sure there’s no way an AI could create the same quality than a human. Everything would sound too similar if not the same. An AI cannot think and cannot feel, besides of that someone had to deliver some input before to make an AI work. So it all depends on the input how good the results can become, but it can’t become better than an “original”. Right? Right!
Suppose it’s a tongue in cheek remark but why would AI use Renoise, or any DAW for that matter? At one point it will just generate the appropriate waveform for the given prompt. Eventually probably even in real time, yes. And no doubt you could make it spit out an XRNS file if you train the model. But why would you?
I haven’t looked into it lately, but last year or so music AI still seems to require human curation. A human that chose which parts work well, and which generated ideas are a dead end. Eventually it will learn that too. And I’m thinking that ‘the people at large’ may actually prefer AI generated music. For the same reason why all the songs on my local radio stations all sound the same. People don’t really want new things. They want more of the same.
edit: lol, I typed that before you posted your message @TNT
But in response to your post. I didn’t say it would be good music, I said it would be generic music. There is nothing magic’al about recreating that formula.
people who feel the need to create will always continue to create, and to some extent, they will create in response to the cultural currents of their time. There’s no doubt (in my mind) that the proliferation of AI will impact many aspects of human life, and the arts are no exception, but creators will continue to make creations. Personally, I think the inbuilt tendency towards genericism of AI will drive people towards greater creativity and towards new forms of expression. Like a creativity arms race. Perhaps the AI will “win,” but don’t we all win if the result is more good music?
If music was my profession, I might feel differently, but as a hobbyist I feel like the musical results are what’s intriguing from a music “consumer” standpoint, and the creative process is what’s (mostly) intriguing from a production and composition standpoint. Like, people still play chess and love it and devote their lives to it, even though computers are better than any human player. Music, being an open (and essentially infinite) system is a bit of a different story, of course…
What I’m really curious to see is if AI can make us feel different (“good”) things musically than human composers can, just by virtue of the structure/nature of their process
also, those mashleg mixes are hilarious
The whole “how to use them properly” argument is a moot point if you train it on whatever your idea of ‘proper’ is.
The same goes for ‘feelings’. Throw enough resources at it and you can teach it that “this sound (tone, timbre & texture) evokes this mood. that sound feels like that and so on”. Then the AI can draw on that with your prompt.
Yes, whatever it makes is based on existing music. But, so is all of our music. We’re all standing on the shoulders of giants. The music we listened to in our lives shapes whatever we make. Both deliberately and subconsciously.
A key difference indeed, the AI will not experiment in the way humans do, no. We occasionally find new sounds/genres by (accidentally or deliberately) doing something “wrong” .
But whether or not AI can do things “better” than an original? First off, what is “better” in this context? AI is probably more effective at figuring out what exactly makes a hit song. The thing that has the exact right mood, tempo and sound for whatever is popular at the time. But it’s based on what’s already out there, indeed.
Though if you have enough resources at your disposal you can make an experimental AI as well, I suppose.
One could make AI to spit out Renoise xrns or train it to even make songs with mouse and keyboard - but why? why not…idk - probably somewhere someone has enough time, brains and money to do it and maybe they will.
Yes, I agree - there will be a lot of generic stuff, there is a lot of generic stuff already. Sorry, but lots of House and Techno is - and I don’t mind listening to some house or techno now and then.
I think the OP post goes ar has to go deeper - eventually the AI songs will sound authentic, good and edgy enough to pass for a human composer or performer or DJ - they can include all kinds of ‘humanizing’ tweaks.
But at some point personally I stop caring about it all - because why would I. Perhaps I am and old school hippie, and stuck in classic rock understanding of stuff - but I want to see humans doing music, playing or composing.
I remember in George Orwel’s 1984 there was a device mentioned called Atomator or Combinator or something - it created songs and lyrics for the proles to enjoy.
Wasn’t that the versificator?
I think @slujr is right in that people who feel the need to create will continue to do so. And I think that @Achenar is spot on in showing the human side of whatever art you will be an advantage in finding an audience.
The thing that kind of killed Rock/Grunge/(Nu)Metal as a mainstream genre is that it got homogenized with the advent of pro-tools and auto-tune.
Earlier this year I saw Sungazer live. It had been a long time since I saw a band, rather than an electronic music performance. And I’ve seen some kickass electronic acts in recent years! Fantastic sound, fantastic shows, fantastic crowds. But the energy of 4 skilled guys on stage playing together…That’s something else. I had forgotten how good live music can be.
Yes, I think Versificator it was. Here’s an old article from Wire: https://www.wired.com/2008/02/does-the-techno/
“Here were produced rubbishy newspapers containing almost nothing except sport, crime and astrology, sensational five-cent novelettes, films oozing with sex, and sentimental songs which were composed entirely by mechanical means on a special kind of kaleidoscope known as a versificator.”
“The tune had been haunting London for weeks past. It was one of countless similar songs published for the benefit of the proles by a sub-section of the Music
Department. The words of these songs were composed without any human intervention whatever on an instrument known as a versificator. But the woman sang so tunefully as to turn the dreadful rubbish into an almost pleasant sound.”
Apologies for the total tangent, but that second quote is the beginning of my favorite part in that whole book
The tune had been haunting London for weeks past. It was one of countless similar songs published for the benefit of the proles by a sub-section of the Music Department. The words of these songs were composed without any human intervention whatever on an instrument known as a versificator. But the woman sang so tunefully as to turn the dreadful rubbish into an almost pleasant sound. He could hear the woman singing and the scrape of her shoes on the flagstones, and the cries of the children in the street, and somewhere in the far distance a faint roar of traffic, and yet the room seemed curiously silent, thanks to the absence of a telescreen.
She knew the whole drivelling song by heart, it seemed. Her voice floated upward with the sweet summer air, very tuneful, charged with a sort of happy melancholy. One had the feeling that she would have been perfectly content, if the June evening had been endless and the supply of clothes inexhaustible, to remain there for a thousand years, pegging out diapers and singing rubbish. It struck him as a curious fact that he had never heard a member of the Party singing alone and spontaneously. It would even have seemed slightly unorthodox, a dangerous eccentricity, like talking to oneself. Perhaps it was only when people were somewhere near the starvation level that they had anything to sing about.
He would have liked to continue talking about his mother. He did not suppose, from what he could remember of her, that she had been an unusual woman, still less an intelligent one; and yet she had possessed a kind of nobility, a kind of purity, simply because the standards that she obeyed were private ones. Her feelings were her own, and could not be altered from outside. It would not have occurred to her that an action which is ineffectual thereby becomes meaningless. If you loved someone, you loved him, and when you had nothing else to give, you still gave him love. When the last of the chocolate was gone, his mother had clasped the child in her arms. It was no use, it changed nothing, it did not produce more chocolate, it did not avert the child’s death or her own; but it seemed natural to her to do it. The refugee woman in the boat had also covered the little boy with her arm, which was no more use against the bullets than a sheet of paper. The terrible thing that the Party had done was to persuade you that mere impulses, mere feelings, were of no account, while at the same time robbing you of all power over the material world. When once you were in the grip of the Party, what you felt or did not feel, what you did or refrained from doing, made literally no difference. Whatever happened you vanished, and neither you nor your actions were ever heard of again. You were lifted clean out of the stream of history. And yet to the people of only two generations ago this would not have seemed all-important, because they were not attempting to alter history. They were governed by private loyalties which they did not question. What mattered were individual relationships, and a completely helpless gesture, an embrace, a tear, a word spoken to a dying man, could have value in itself. The proles, it suddenly occurred to him, had remained in this condition. They were not loyal to a party or a country or an idea, they were loyal to one another. For the first time in his life he did not despise the proles or think of them merely as an inert force which would one day spring to life and regenerate the world. The proles had stayed human. They had not become hardened inside. They had held on to the primitive emotions which he himself had to re-learn by conscious effort. And in thinking this he remembered, without apparent relevance, how a few weeks ago he had seen a severed hand lying on the pavement and had kicked it into the gutter as though it had been a cabbage-stalk.
Don’t you think one day people will pay some real hot bucks to get some good music made by an actual human who is still able to do it? Like this: “Sheesh…want some real hot beats from the 90s? Still fresh, only little crackles… But don’t tell anyone you’re listening to it. You know, it doesn’t have that mind manipulation in it like the new shit, it will make you feel all giddy and dance instead of comply & obey…”
And don’t forget, they need a lot of junk music, to train an AI to make even more junk music that sounds just like the crap they’ve trained it with. Just with some extra weirdness because AI only mimics what it knows and doesn’t understand it intellectually. Who will make all the training tracks for that new style of music that is the new fad?
But those nasty 1984 visions will now become reality in any state ruled by angry people, which probably means the book failed because it was written to warn of such things so they never happen… And we should not forget that, even when it is basically a proper salvation for most of them, now a lot of people who made jingles or pointless soap opera music scores will now go jobless and maybe have to teach people, how they could still make music without using AI…
Yes - eventually the AI glory will fade in its generative sameness. Yes, it will be a shift and paradigm change etc etc… but eventually… music comes from life, from experiences, the good, bad and ugly of existence and universe.
I am feeling optimistic.
There’s a poet I follow, Anthony Etherin, who works with really tight constraints. There are times when it feels like the poet isn’t saying anything with the words of the poem, but instead is showing the reader something about the language in light of the constraints.
Right now, the output of LLMs is a little like that.
The more I think about this, the more right I think you are @Achenar . Though not necessarily because of the possibility people will assume whatever you show off is made by AI. A lot of the videos I watch on youtube are people demonstrating a (new) thing, synths mostly, in my case. I’m there for the synth/technique/whatever but I’m hearing their music. Not everyone’s music I’m exposed to that way is my cup of tea. But there is also a few of those people who have actually turned me onto their music by showing their process.
So because of your video I decided to give it another try to do exactly that, showing part of the process. And I plan to stick with that approach for a while. Even though filming myself is quite a hassle. In an act of shameless self-promotion, let me post my video here.