Something interesting... the first 10 seconds or so of the "Death Growl" example[1] is basically copied verbatim from "Ov Fire And The Void" by Behemoth.
More specifically, I think the part that seems copied is at 2:13 of the original[2], as it leads into a solo-ish bit which in the AI version sounds similar still, but goes on to do its own thing:
> Additionally, our memorization-effect experiments in Section 11 demonstrate that our design maintains creativity without plagiarizing, even under strong training set conditioning.
That decision was ridiculous. It's pretty obvious that the Robin Thicke song is a $1.50 Great Value version of "Got To Give It Up" because of the aesthetic similarities but they have nothing to do with each other melodically or harmonically... "Blurred Lines" sounds like I V with a walk at the end whereas "Got To Give It Up" is more like a I IV V. The vocal melodies aren't the same nor is the bass. They have different arrangements. The percussion isn't the same.
The only things they have in common are vibes (in the contemporary sense, not vibraphones). Two dudes singing about sex in falsetto at 120bpm over prototypical R&B/funk elements isn't special. If that's the bar for copyright infringement then 99% of the popular music canon is illegally-derivative. Marvin Gaye was a singular talent but that doesn't mean that his heirs should be able to collect money every time somebody plays an electric piano bassline and sings about making whoopie in alto II.
Not using streaming services help somewhat, but I'm not looking forward to having to vet artists I come across for whether they did any substantial original thinking or work for themselves. Tired of snake oil foolishness.
Very nice. Anyone know of projects that aren't tackling the full-song problem but rather instrument parts/loops/stems/acapellas? I'd like something that's more like "infinite AI Loopcloud/Splice" most of these full-song models don't do well to be asked for individual parts in my experience (though I will have to try it with this one).
This gets discussed a lot but unfortunately there's just not much out there around this.
The closest thing I've seen is virtual drummers in Logic X which will follow along with the structure of your song and generate a percussive accompaniment. It's no substitute for a real drummer but it's serviceable.
Yeah. Or like, a loop that plays continuously and has style parameters exposed you can tweak with a controller like a Midi Fighter Twister and get feedback from in real-time. Then you could do something akin to DJ/live production by having two of these going in sync with each other into a mixer. (Tweak params of the cue track until you like it, transition at a phrase point, repeat).
Yesssss. Something I'd really like is some kind of virtual drummer pedal with an expression pedal (pedals?) so that it can be tweaked in real time.
One of the most fun things about jamming with other musicians is that things change with the ebb and flow of the band's energy. i always miss it when I'm just fooling around at home with a looper.
What is the use case for music generation models? I see usecases for alot of the other foundation models like text, image, tts, sst, but why do I want AI generated music?
It's the only way for Spotify to turn a profit without all the human work of having to scout, sign and promote flesh-and-blood artists, the things that real music labels do.
I’ve mostly used them for laughs with my friends. Sometimes generating “custom” songs with funny lyrics, but most fun so far is editing lyrics of existing songs to say ridiculous things for fun.
No real clue how someone would use them for a more serious endeavor, only thing I could imagine would be to quickly iterate/prototype with song structures on a fixed seed to generate ideas for a real composition. Consider the case of an indie game developer or film maker getting some placeholder music to test the experience during early throwaway iterations.
An actual serious answer is to help musicians brainstorm while writing. It's so good at helping me come up with ideas, or converting an idea to another genre.
If the computer is fast enough to keep up with real time music, one could imagine a model that can jam with you with other instruments, reacting to and complementing what you're doing.
I'd imagine that would not get much traction, as there are probably people that would pay you to do that. Since people pay to listen to music, and not to make it, AI has much more utility there. The great thing is, if you are someone who gets joy in making music, AI in no way limits this.
The idea that all music is this super special blood sweat and tears spiritual thing is bullshit. Pop music is written to sell money and this doesn't prevent artists from doing their thing. Relax and let people like things.
It doesn't. But fast food and pre-cooked meals made it so a large percentage of people can't cook with raw or unprocessed ingredients anymore and can only rely on giant companies to feed themselves.
Junk food is OK every once in a while, but if that's all you eat you're not going to be healthy. I believe the same is true for the media you consume.
Edit: Also, to add to your analogy, junk food made it so people's palates became accustomed to high amounts of sugar, fat and salt, to a point where some people "Don't like water". If you've only known pizza rolls and chicken tenders, you're unlikely to enjoy a chicory salad or a even a moussaka. And if all your entertainment is call of duty, marvel movies and AI generated music, you're unlikely to enjoy Weather Report, Rachmaninoff or Pynchon. I'm not saying that to to be elitist, but we've got to stop dumbing down culture and entertainment
This is elitist bullshit. I've been a musician my entire life and made a decent living doing it. People can choose to listen to whatever they want and it's not your job to police what kind of music people can or can't make or what people can choose to listen to.
Something interesting... the first 10 seconds or so of the "Death Growl" example[1] is basically copied verbatim from "Ov Fire And The Void" by Behemoth.
More specifically, I think the part that seems copied is at 2:13 of the original[2], as it leads into a solo-ish bit which in the AI version sounds similar still, but goes on to do its own thing:
[1] https://map-yue.github.io/music/moon.death_metal.mp3
[2] https://youtu.be/vAmnsKKrt9w?t=133
> Additionally, our memorization-effect experiments in Section 11 demonstrate that our design maintains creativity without plagiarizing, even under strong training set conditioning.
https://arxiv.org/html/2503.08638v1#S11
Funny because since the Blurred Lines lawsuit you can be infringing for using the same chord progression.
That decision was ridiculous. It's pretty obvious that the Robin Thicke song is a $1.50 Great Value version of "Got To Give It Up" because of the aesthetic similarities but they have nothing to do with each other melodically or harmonically... "Blurred Lines" sounds like I V with a walk at the end whereas "Got To Give It Up" is more like a I IV V. The vocal melodies aren't the same nor is the bass. They have different arrangements. The percussion isn't the same.
The only things they have in common are vibes (in the contemporary sense, not vibraphones). Two dudes singing about sex in falsetto at 120bpm over prototypical R&B/funk elements isn't special. If that's the bar for copyright infringement then 99% of the popular music canon is illegally-derivative. Marvin Gaye was a singular talent but that doesn't mean that his heirs should be able to collect money every time somebody plays an electric piano bassline and sings about making whoopie in alto II.
The youtube link is suddenly not available any more (at least in the UK)
Does Shazam think it is the same?
Not using streaming services help somewhat, but I'm not looking forward to having to vet artists I come across for whether they did any substantial original thinking or work for themselves. Tired of snake oil foolishness.
Very nice. Anyone know of projects that aren't tackling the full-song problem but rather instrument parts/loops/stems/acapellas? I'd like something that's more like "infinite AI Loopcloud/Splice" most of these full-song models don't do well to be asked for individual parts in my experience (though I will have to try it with this one).
This gets discussed a lot but unfortunately there's just not much out there around this.
The closest thing I've seen is virtual drummers in Logic X which will follow along with the structure of your song and generate a percussive accompaniment. It's no substitute for a real drummer but it's serviceable.
Try https://magenta.withgoogle.com/infinite-crate
https://suno.com/studio-waitlist Just a waitlist so far, but looks like this is the direction suno is going
Yeah... I hope this is what their plan is with that, but I'm not entirely certain.
Also live AI dueting would be interesting, like having a virtual guitarist you could jam/duet with.
Just saw this today: https://x.com/jesseengel/status/1953496623696556478
Yeah. Or like, a loop that plays continuously and has style parameters exposed you can tweak with a controller like a Midi Fighter Twister and get feedback from in real-time. Then you could do something akin to DJ/live production by having two of these going in sync with each other into a mixer. (Tweak params of the cue track until you like it, transition at a phrase point, repeat).
Yesssss. Something I'd really like is some kind of virtual drummer pedal with an expression pedal (pedals?) so that it can be tweaked in real time.
One of the most fun things about jamming with other musicians is that things change with the ebb and flow of the band's energy. i always miss it when I'm just fooling around at home with a looper.
Like this? https://aistudio.google.com/apps/bundled/promptdj?showPrevie...
Oh! The idea that each parameter could be associated with a text style prompt is a refinement of the idea for sure.
What is the use case for music generation models? I see usecases for alot of the other foundation models like text, image, tts, sst, but why do I want AI generated music?
Now you don’t need to know how to make music! You’re finally free of all those pesky, elitist musicians gate-keeping music!!!!1!
It's the only way for Spotify to turn a profit without all the human work of having to scout, sign and promote flesh-and-blood artists, the things that real music labels do.
I’ve mostly used them for laughs with my friends. Sometimes generating “custom” songs with funny lyrics, but most fun so far is editing lyrics of existing songs to say ridiculous things for fun.
No real clue how someone would use them for a more serious endeavor, only thing I could imagine would be to quickly iterate/prototype with song structures on a fixed seed to generate ideas for a real composition. Consider the case of an indie game developer or film maker getting some placeholder music to test the experience during early throwaway iterations.
Games?
Generating crappy background music for reality TV?
Yeah, this seems the most likely. Just loads of royalty free background music for industries that want that.
Put more slop on Spotify hoping it will generate some amount of revenue.
An actual serious answer is to help musicians brainstorm while writing. It's so good at helping me come up with ideas, or converting an idea to another genre.
That's called 'jamming' and it's been possible for musicians since the time cavemen figured out they could hit sticks on things to create a rhythm.
If the computer is fast enough to keep up with real time music, one could imagine a model that can jam with you with other instruments, reacting to and complementing what you're doing.
thats basically existed at least since 1990 with band in a box
yeah, but have yall made any progress in a model that can have sex with my partner for me?
I'd imagine that would not get much traction, as there are probably people that would pay you to do that. Since people pay to listen to music, and not to make it, AI has much more utility there. The great thing is, if you are someone who gets joy in making music, AI in no way limits this.
[flagged]
It does seem capable of stirring a lot of emotion, if your comment is any indication.
The idea that all music is this super special blood sweat and tears spiritual thing is bullshit. Pop music is written to sell money and this doesn't prevent artists from doing their thing. Relax and let people like things.
This is the fast food of music. It's soulless, and steals from real artists.
Fast food doesn't steal from gourmet restaurants. If it's so bad, who cares?
It doesn't. But fast food and pre-cooked meals made it so a large percentage of people can't cook with raw or unprocessed ingredients anymore and can only rely on giant companies to feed themselves.
Junk food is OK every once in a while, but if that's all you eat you're not going to be healthy. I believe the same is true for the media you consume.
Edit: Also, to add to your analogy, junk food made it so people's palates became accustomed to high amounts of sugar, fat and salt, to a point where some people "Don't like water". If you've only known pizza rolls and chicken tenders, you're unlikely to enjoy a chicory salad or a even a moussaka. And if all your entertainment is call of duty, marvel movies and AI generated music, you're unlikely to enjoy Weather Report, Rachmaninoff or Pynchon. I'm not saying that to to be elitist, but we've got to stop dumbing down culture and entertainment
This is elitist bullshit. I've been a musician my entire life and made a decent living doing it. People can choose to listen to whatever they want and it's not your job to police what kind of music people can or can't make or what people can choose to listen to.
At least "Fast food doesn't steal from gourmet restaurants."