I hate to admit it but... AI music?

KingDuppy · 2024-02-10T08:48:26.235Z

It’s an interesting question, what’s the difference between an AI song, a mash-up, a cover, or even an original song that’s heavily derivative of other music?

I think the difference is that when a person is inspired by music they hear, even when they do a cover or bad copy of a song, they bring more than other musical input to it. They bring their whole lives.

Say someone in 2024 tries to make a Motown song from 1964—they are doing a lot more than merely copying the sonics and timings.

The person hearing the song in 2024 has 60 years of cultural remove, they live in a different time, probably a different place, and have different concerns and fears and hobbies, different beliefs, that all inform the song they make.

An AI has none of this context. It has mountains of inputs, but it cannot truly understand them the way a human does, because it has no body and no life story. To an AI, music may as well be a dress pattern.

But the musician making the shitty cover, they have friends, family, rent to pay, a whole story that’s uniquely theirs and changes the song they end up making, for good or bad.

A computer can mimic a mind, but our minds are not computers. I think comparing the two does a disservice to the human brain, a marvel of nature, and the people doomed to carry the damn things around in our heads all day.

LCVL · 2024-02-10T08:55:15.131Z

KingDuppy:

An AI has none of this context.

Like a “homo abilis”?
And what if it’s just a matter of giving AI enough time (and data)?

Not trying to defend AI, just speculating…
After all we can all agree this is just the very, very beginning…

KingDuppy · 2024-02-10T09:01:00.633Z

No, not like one at all, because homo habilis was an animal that got smarter, not a machine made to mimic a smart animal.

I think humans ascribe way too much value to our “intelligence” because it flatters us and makes us feel different from the other bloodthirsty creatures. But a dog is exactly as smart as a dog needs to be.

KingDuppy · 2024-02-10T09:04:26.200Z

What I mean is, the homo habilis still has context. And a body. It gets hungry and scared. It lives in a pack that has its own dynamics

Humans weren’t automotons before we got smarter. And lots of non-human animals have cultures.

cema · 2024-02-10T09:15:29.482Z

I recently heard a club owner who wants an AI-DJ to play music to the clubbers.
Obviously for saving cost on gig fee.
The interesting question whether or not the audience accept AI or looking for the human interaction. Even for electronic music DJing I really enjoy to see how the DJ is working. If vinyl that adds another layer of “magic”.
However I think AI is not just a hype, the possibilities are endless and the progression is amazing.
Imagine a back to back set with Richie Hawtin and his AI-clone.

iamsaitam · 2024-02-10T09:25:10.028Z

Degrees matter though, don’t they? The comparison should be with artisans and their workshops crafting products vs the industrial revolution. AI isn’t simply another tool it’s a means of infinite production and reproduction of existing tendencies.

Soon enough the big labels will just automate the whole process and hire “taste makers” to find the next hits. Then they’ll ride the cool wave and create “artificial”/digital personas that will do digital signings, interviews, live gigs (with human musicians) and the capital will continue to go upwards and concentrate further in fewer hands.

JPM · 2024-02-10T17:21:48.851Z

What about this.
EvilCorp uses AI to create trillions of songs a day, including lyrics, melodies etc. then copyrights it all.
Now no one can make any music because it’s all copying them

DaveMech · 2024-02-10T21:08:49.465Z

iamsaitam:

The comparison should be with artisans and their workshops crafting products vs the industrial revolution

Right and they still exist, don’t they ? hand made stuff is still being made, sold and bought.

I get your point of course, but the part of the industry you’re talking about is the main stream and it’s not that much different from its current form in that the people at the top get most of the money, radio stations play the playlists that are curated for them and to cater to the masses to sell more and so on. AI will definitely automate a lot of that. Spotify will have (already has I believe) AI generated music playlists etc.

It’s pretty bleak. However, it won’t completely substitute humans performing music or creating music. And it definitely won’t substitute people loving to see and hear actual humans play music. It will co-exist. And it will definitely also be used as a new way of generating ideas or as a tool to create completly new forms of music from which is already evident.

It for sure has dark sides to it as with any revolution, but I don’t think it will be the downfall of human creativity or anything like that.

waftlord · 2024-02-10T21:32:39.390Z

We maybe in the last batch of humans hand making music before it’s all amalgamated into “the model” and people just resort to generate prompted deviations of everything recorded pre 2030.

Creative prompting is the new art.
Preative crompting.

DaveMech:

but I don’t think it will be the downfall of human creativity or anything like that.

ViolentMeals · 2024-02-11T18:16:37.615Z

KingDuppy:

the human brain, a marvel of nature, and the people doomed to carry the damn things around in our heads all day.

I love this…

Leonsarmiento · 2024-02-11T18:47:32.941Z

AI for me it’s just pretty convenient:

instead of go out and hunt down drum breaks (that later will be mangled - dissected and re arranged back again) I just prompt my Google Colab for them.

It’s just like an arpeggiator or a VST : they make your life easy… But your still have to do your part and plan what you want to do with these resources.

Of course, you can also hotknob AI generated music like Splice loops and call it a day.

Leonsarmiento · 2024-02-11T18:52:43.869Z

waftlord:

Creative prompting is the new art.
Preative crompting.

“Prompting is sooo 2023…” It’s the next one.

The future is just “think and imagine what you want. The machine will understand”:

images (21)739×415 20.8 KB

waftlord · 2024-02-11T18:57:07.294Z

if i had access to a decent image creation llm i’d prompt a mash up of boobs bums bacon sandwiches and machinedrums to convey how i’d be terrified and ashamed to see accurate results direct from my mind…

“please filter out all the animalistic instincts and obsessive intrusions” will be a required checkbox

Edit: actually that maybe exactly what I’m striving for in sounds i create….

Rabban · 2024-02-11T19:24:36.730Z

I gave it a shot. Using the musicgen large model I generated a bunch of desert trance clips/songs. The results are very interesting. The words you use in the prompt can produce wildly different results.

I went for 3.5 minutes of output in each clip, and perhaps that was a bit much. I’m interested to see how Octatrack slices these up, and perhaps I should have made them shorter. Using the large model spiked around 9GB GPU memory, which really is not much compared to big LLMs or image/video generation.

I see a big future for this type of thing and I’m surprised Ableton is not leading the way. Look at adobe with Photoshop. You take a picture of a lake and then draw a circle in the middle and say “add some ducks” and it gives you a bunch of realistic options to paint in the ducks. This should be seamless in Ableton where you highlight an area and say “Give me a syncopated drum fill here” and it inserts a context aware clip… I.e., not some random drum sample but one tailored to what’s going on in the timeline. Maybe eventually.

DJSpaceP · 2024-02-11T19:56:28.315Z

AI music generation is improving all the time. There’s a service called suno.ai that can generate surprisingly decent songs, even with vocals that don’t suck if you know how to prompt and “lead” it from part to part. But there are still many of the audible artifacts, glitches in places, etc. But better sounding than most I’ve heard.

Right now, AI is most useful to me in terms of vocals. For instance, in my latest track, I used AI to generate a vocal track from my lyrics, and did several dice rolls on each part of the vocal until the melodies and timings were close to where I wanted.

Then I brought the parts into RipX where I could clean up and fix melodies and phrasings that weren’t right.

I used that “fixed” vocal in my Ableton project with all of the instrumentation I created. Once I had sliced and diced the vocals even more and added some effects, I had a pretty good track.

But the voice track was still of lower quality than I wanted, so with the sliced and diced vocal stem from Ableton, I used ANOTHER AI service (audimee.com) that can transform a vocal (real or AI) into any of their dozens of virtual singers. I found one of their virtual singers that sounded “right” for the track and had the AI effectively duplicate the uploaded original AI track, but with better quality vocals.

Then I layered the source AI vocal and new AI vocal together and with a few more fixes (weird consonant mismatches, etc), and it sounded great.

I didn’t have to sing a single thing (been dealing with a flu for the last two weeks) and got a vocal that fits my track, plus the services and apps I use are copyright free, trained on open sources or using training data from singers, etc. who were properly licensed and paid for their work.

Just be emotionally prepared for the day coming soon when you can prompt the AI with, “make me a finished club banger in the style of _________ meets __________”, and within minutes you can download a polished track ready to release.

Leonsarmiento · 2024-02-11T21:06:15.389Z

Just imagine YouTube ads delivered directly into your brain.

Unskipable publicity. Unavoidable consumer tracking.

Horrible. AI loops is the less harmful thing.

waftlord · 2024-02-11T21:15:16.576Z

the only exciting thing to me is SPEED. adverts will be there, but we’ll normalise ignoring and swipe aside as we do now.

but if i can make music how i perceive it by looking at things and touching my fingers then that could be really fun. same with coding. really interested if it would be quicker than a keyboard. still not clear to me at this stage.
minority report style interfaces will be here within a year…

Leonsarmiento · 2024-02-11T22:26:51.894Z

waftlord:

if i can make music how i perceive it by looking at things and touching my fingers then that could be really fun.

100% agree

skinpop · 2024-02-12T09:15:01.045Z

Not only that, but we also tend to ascribe too much value to instrumental reasoning vs substantive/value reasoning. It’s how you get incredibly well educated STEM people with degrees from MIT or whatnot who think Jordan Peterson is the best philosopher of our age.

harmonyinultra · 2024-02-12T09:20:37.320Z

I’m loving Lee Gambles latest rec, ‘models’. It’s full of disembodied ai voices.