I mentioned this in another comment thread, recently, but I value art, including music, for the fact that another human being(s) created it to express something. I have zero interest in it if it's AI-generated. The thought of endless content created from the void for me to gorge on like oreos fills me with nothing but despair.
I think AI will raise the bar for all kinds of artistic mediums. There’s already more music, literature, film etc than you can possibly consume - if you can easily make middling art with a prompt then it’s going to be swamped by all the other stuff, the really good, unique, fresh art will rise to the top. Perhaps they’ll use AI but it’ll still be the same amount of work put into and the quality will be better. I think nobody is going to get rich and famous from low-effort promoting.
Survivorship bias. You can't tell at some point. You probably already are listening to AI songs, and looking at AI images. You can only tell what you know. Given the amount generated daily its more probable than improbable.
With AI it’s very easy to create generic, common Pop or any other formulaic garbage music. It might be a good thing for the music industry, most of the low effort might just vanish in the sea of AI gen. This may actially raise the bar for music in general while at the same lowering the bar to simply prompting which is just a stage that will go away.
This is not a pro ai comment, I hope some bios bug bricks all the nvidia rtx gpus in the planet.
With that said... ai generated music, the same can be said about software... but compilers are basically ai, there is no way a for loop on extremely high level abstraction that then open connections to some service, generates jtokens, validates, and apply math non stop from a couple of lines is anything but alien to the 99% of us (yes I know you can write assembly for all your programs and you are better than the compiler)
What if you say, if humans where not involved and a phisically non electric instrument where involved and done with only humans, it's AI assited? because again you don't understand ALL the relationships, how it's designed every single thing, all the things in a music studio to master a song, ALL the related mechanisms used in the promotion, distribution and live execution of those songs... and yes, again there is some of you who can build all those things from sand and pumped from the ground with a tool you designed bare petroleum and program, design and execute EVERY SINGLE THING required to get a record of music you like...
So if a group of 20 people spends 20 weeks working with AI tools to get you a song, then you don't care I guess right?
Did you know that Enya for this song recorded thousand something times voices to make a chorus until it had a particular form... does that mixing of thousands of executions and the tools to analyze and give it the form wanted qualify as AI?
If you move sliders and push buttons and you don't understand the math behind it, are you vibe mixing a song?
Yes, if it is a "give me a catchy song with lyrics that talk about big butts" and then publish... yes this is crap.
It's really depressing that I have gotten into music composition right when this is happening.
All of these projects are basically just destroying entire fields of human creativity.
It doesn't seem so bad at first when the quality is mediocre, but as soon as they get good, then any human composition will be completely devalued, despite the enormous amount of knowledge, skill and effort that goes into producing one.
A composer who writes a piece of music themselves, instead of being recognised and admired basically becomes the equivalent of someone who bakes their own sourdough instead of going to the supermarket: a curiosity, someone who does things the hard way for their own amusement.
Graphic artists and bloggers were the first to suffer this, but it feels like it's just going to happen to basically every type of creative expression that can be represented digitally. It's just a question of how many months until that happens.
I'm also saddened by the impact of tech on creative industries. But art always was something you do for yourself and to share with others, for some fundamentally human experience. Could an AI pray for you or ride your bike? I don't know anyone in my personal life who wants to listen to AI music, view AI art, or read AI books. I wish art careers were easier, but I don't think a career alone is the right reason to make art.
Much like the restaurant industry, creative industries will probably bifurcate. Cheap (AI, drivethrough, takeout) for daily and most use cases, experiential (vinyl records, concerts, upscale restaurants) for special occasions, connoisseurs.
Who cares? The world is being handed an infinite creation machine that conjures entire worlds into existence, radically reshaping the future of the species and what our descendants will be able to do, and your reply is ~"they took our jobs!"
Figure out how to stand on the shoulders of giants like any of your predecessors of note.
You are free to perform create expression, I love writing blogs, my friends like reading and talking about the blogs with me, I dont care if Claude or GPT can write much better blogs. In the same way if it was just about creative expression, there would actually be 0 problems with AI-generated music, you would keep on making and listening to music you like, and you wouldn't care, but usually its not it, its people who want to make money, get famous, tie their value to the product that complain about this.
> you would keep on making and listening to music you like, and you wouldn't care
It becomes harder to find and therefore listen to music you like when every avenue of discovering music is drowned out with 1000 AI-generated songs for every 1 human-made song.
> people who want to make money,
Is it such a bad thing to want to be able to pursue your creativity full-time? For creative fields that require thousands of hours of practice to achieve results other people will enjoy, should we reserve such skilled expression only to those born wealthy enough to not need to work?
> It becomes harder to find and therefore listen to music you like when every avenue of discovering music is drowned out with 1000 AI-generated songs for every 1 human-made song.
Creative expression and consumption of media are different questions. You are right about that, but that's a separate issue.
> Is it such a bad thing to want to be able to pursue your creativity full-time? For creative fields that require thousands of hours of practice to achieve results other people will enjoy, should we reserve such skilled expression only to those born wealthy enough to not need to work?
Welcome to capitalism, enjoy your stay. And if you dont like it, you can always wait for the coming times when informational work is completely automated, which means that traditional social mobility completely disappears, and capital is locked/frozen forever and we will have serfdom instead. So I would recommend you enjoy the present.
It's being handed an infinite slop machine, drowning out genuinely good content and making it harder than ever to discover. As it becomes harder to discover, fewer talented creatives are able to make a living. We are destroying our own culture with a machine that enforces regression to the mean on creativity.
I've been waiting for symbolic music generators for ages. This is so nice for musicians. Being able to intelligently translate parts, generate additional parts, generate sheet music from audio, generate variations on pieces, making pieces easier or harder.
The vocals in the few songs I listened to are so very clearly AI. I find Suno's much more credible in many genres.
If anyone has tried Suno's Studio 2, I would love to hear your take on it. At $24-30/month, it's approaching Ableton Live 12 Suite rent-to-own territory, without the "own" part.
For an informed take on where this is going for musicians and the music industry, I really recommend Adam Neely's video essay "Suno, AI Music, and the Bad Future": https://www.youtube.com/watch?v=U8dcFhF0Dlk
I say this as someone with no skin in the game, but I don’t think it’s necessary for AI to replace human creativity in music. Instead, it can be a tool for augmentation.
There has almost always been backlash against new technologies in music. Ultimately, musicians adopted those technologies and continued producing great works.
I recently dug some 20-year-old CDs out from storage that were full of riffs and chord progressions for songs I never got around to completing. I fed them to Suno as a sort of “what might these have sounded like if I’d finished them?” experiment. I found the results quite impressive and plan to publish them some time in the near future.
None of the tracks sound like marketable products to me, and I certainly don't think AI is going to generate the next Bohemian Rhapsody -- but it definitely could be a great thing for curing writer's block.
It's interesting that the “Classical Music” sample ended up with a style prompt of “Contemporary Christian”.
It's even more interesting that the “Contemporary Christian” music it generated was specifically Mormon music (specifically about the exile of Lehi and his family from Jerusalem into the wilderness, during which they reach the ocean, build a boat, and sail to the Americas), but in the style of contemporary Christian rock.
The majority of people, I think, don't want to listen to AI music or view AI art. Systems like this will produce muzak for ubers, cafes, phone queues, advertisements. Some artists will make hyperreal work by sampling it. But to listen to these "songs" willingly is not something I expect most people will choose to do.
I went and listened to a bunch of the examples and it feels like it's trained in one very contemporary style. E.g. I hear a vaguely buildup-drop pattern that sounds a lot like EDM in places I wouldn't expect it.
I really want to like ai music but none of them seem to be trained on reward models that reward "ambiance" or any sort of interesting sound design (which this paper obviously doesn't even concern), which might be reflective of the people training them not having niche music tastes
"The Petty and the Small;
Are overcome with gall ;
When Genius, having faltered, fails to fall.
Klapaucius too, I ween,
Will turn the deepest green
To hear such flawless verse from Trurl's machine."
That example "Comedy" sounds like the most soulless synthetic example of "music" ever. Like the music analogue of a PR-speak article of a big corpo generated by an LLM. It's still technically impressive but it's 0/10, absolutely empty of soul.
All AI generated art feels somewhat masturbatory to me, but AI generated music is the most baffling of them all. It's like what if someone thought Muzak had too much soul.
You know how they used heavy metal music to break the spirit of Manuel Noriega? I'm sure there's some DoD (or is it DoW now?) psychological operations people right now studying the prolonged effects of high volume mediocre AI slop music.
The playlist included "the sound of jet planes all night" and "the screams of rabbits being slaughtered". I don't think they really needed AI for that, anything loud would probably do the trick after 51 days.
The use case is control. It's essentially a controlnet for the actual music writing. So a musician could compose melody, harmony, and lyrics and generate many performances of the same song until they get one they like. A non-musician could similarly generate the symbolic input from scratch, and perhaps they like the composition but not the performance. This allows them to re-render and re-render until they like the performance.
> a musician could compose melody, harmony, and lyrics and generate many performances of the same song until they get one they like
Can't relate to this. Art doesn't come out as a left-brained problem solving exercise or identifying the right product for a market. It's about expression. You do it when you're feeling things.
I'm a composer and I can 100% get behind this workflow. I will often compose on one instrument (guitar or keys) and then throw the midi into a DAW and experiment with sounds.
I'm not species-ist. If a digital sentient being has something to say and they can express it through music, I'm more than willing to listen.
But listening to these samples, it's obvious that, while technically proficient, they have no voice. Funnily enough, neither does most of the human produced music slop that existed before gen AI.
All it exposes is that most people have no taste. This is as good as shuffling through popular songs on Spotify.
Am I as good as Claudio PRC, Donato Dozzy, Ken Ishii or Takkyu Ishino? Not even close. But wanting to make more is reward enough for me to keep trying.
I mentioned this in another comment thread, recently, but I value art, including music, for the fact that another human being(s) created it to express something. I have zero interest in it if it's AI-generated. The thought of endless content created from the void for me to gorge on like oreos fills me with nothing but despair.
I think AI will raise the bar for all kinds of artistic mediums. There’s already more music, literature, film etc than you can possibly consume - if you can easily make middling art with a prompt then it’s going to be swamped by all the other stuff, the really good, unique, fresh art will rise to the top. Perhaps they’ll use AI but it’ll still be the same amount of work put into and the quality will be better. I think nobody is going to get rich and famous from low-effort promoting.
Survivorship bias. You can't tell at some point. You probably already are listening to AI songs, and looking at AI images. You can only tell what you know. Given the amount generated daily its more probable than improbable.
With AI it’s very easy to create generic, common Pop or any other formulaic garbage music. It might be a good thing for the music industry, most of the low effort might just vanish in the sea of AI gen. This may actially raise the bar for music in general while at the same lowering the bar to simply prompting which is just a stage that will go away.
Absolutely agreed. Art made by humans is interesting not just because it sounds it looks novel, but because it reveals something about the artist.
This is not a pro ai comment, I hope some bios bug bricks all the nvidia rtx gpus in the planet.
With that said... ai generated music, the same can be said about software... but compilers are basically ai, there is no way a for loop on extremely high level abstraction that then open connections to some service, generates jtokens, validates, and apply math non stop from a couple of lines is anything but alien to the 99% of us (yes I know you can write assembly for all your programs and you are better than the compiler)
What if you say, if humans where not involved and a phisically non electric instrument where involved and done with only humans, it's AI assited? because again you don't understand ALL the relationships, how it's designed every single thing, all the things in a music studio to master a song, ALL the related mechanisms used in the promotion, distribution and live execution of those songs... and yes, again there is some of you who can build all those things from sand and pumped from the ground with a tool you designed bare petroleum and program, design and execute EVERY SINGLE THING required to get a record of music you like...
So if a group of 20 people spends 20 weeks working with AI tools to get you a song, then you don't care I guess right?
Did you know that Enya for this song recorded thousand something times voices to make a chorus until it had a particular form... does that mixing of thousands of executions and the tools to analyze and give it the form wanted qualify as AI?
If you move sliders and push buttons and you don't understand the math behind it, are you vibe mixing a song?
Yes, if it is a "give me a catchy song with lyrics that talk about big butts" and then publish... yes this is crap.
It's really depressing that I have gotten into music composition right when this is happening.
All of these projects are basically just destroying entire fields of human creativity.
It doesn't seem so bad at first when the quality is mediocre, but as soon as they get good, then any human composition will be completely devalued, despite the enormous amount of knowledge, skill and effort that goes into producing one.
A composer who writes a piece of music themselves, instead of being recognised and admired basically becomes the equivalent of someone who bakes their own sourdough instead of going to the supermarket: a curiosity, someone who does things the hard way for their own amusement.
Graphic artists and bloggers were the first to suffer this, but it feels like it's just going to happen to basically every type of creative expression that can be represented digitally. It's just a question of how many months until that happens.
I'm also saddened by the impact of tech on creative industries. But art always was something you do for yourself and to share with others, for some fundamentally human experience. Could an AI pray for you or ride your bike? I don't know anyone in my personal life who wants to listen to AI music, view AI art, or read AI books. I wish art careers were easier, but I don't think a career alone is the right reason to make art.
Much like the restaurant industry, creative industries will probably bifurcate. Cheap (AI, drivethrough, takeout) for daily and most use cases, experiential (vinyl records, concerts, upscale restaurants) for special occasions, connoisseurs.
Who cares? The world is being handed an infinite creation machine that conjures entire worlds into existence, radically reshaping the future of the species and what our descendants will be able to do, and your reply is ~"they took our jobs!"
Figure out how to stand on the shoulders of giants like any of your predecessors of note.
Engineers understand that every technology represents inherent tradeoffs.
They’re talking about creative expression, not jobs
You are free to perform create expression, I love writing blogs, my friends like reading and talking about the blogs with me, I dont care if Claude or GPT can write much better blogs. In the same way if it was just about creative expression, there would actually be 0 problems with AI-generated music, you would keep on making and listening to music you like, and you wouldn't care, but usually its not it, its people who want to make money, get famous, tie their value to the product that complain about this.
> you would keep on making and listening to music you like, and you wouldn't care
It becomes harder to find and therefore listen to music you like when every avenue of discovering music is drowned out with 1000 AI-generated songs for every 1 human-made song.
> people who want to make money,
Is it such a bad thing to want to be able to pursue your creativity full-time? For creative fields that require thousands of hours of practice to achieve results other people will enjoy, should we reserve such skilled expression only to those born wealthy enough to not need to work?
> It becomes harder to find and therefore listen to music you like when every avenue of discovering music is drowned out with 1000 AI-generated songs for every 1 human-made song.
Creative expression and consumption of media are different questions. You are right about that, but that's a separate issue.
> Is it such a bad thing to want to be able to pursue your creativity full-time? For creative fields that require thousands of hours of practice to achieve results other people will enjoy, should we reserve such skilled expression only to those born wealthy enough to not need to work?
Welcome to capitalism, enjoy your stay. And if you dont like it, you can always wait for the coming times when informational work is completely automated, which means that traditional social mobility completely disappears, and capital is locked/frozen forever and we will have serfdom instead. So I would recommend you enjoy the present.
But is it stopping creative expression?
It's being handed an infinite slop machine, drowning out genuinely good content and making it harder than ever to discover. As it becomes harder to discover, fewer talented creatives are able to make a living. We are destroying our own culture with a machine that enforces regression to the mean on creativity.
I've been waiting for symbolic music generators for ages. This is so nice for musicians. Being able to intelligently translate parts, generate additional parts, generate sheet music from audio, generate variations on pieces, making pieces easier or harder.
So many possibilities.
The vocals in the few songs I listened to are so very clearly AI. I find Suno's much more credible in many genres.
If anyone has tried Suno's Studio 2, I would love to hear your take on it. At $24-30/month, it's approaching Ableton Live 12 Suite rent-to-own territory, without the "own" part.
For an informed take on where this is going for musicians and the music industry, I really recommend Adam Neely's video essay "Suno, AI Music, and the Bad Future": https://www.youtube.com/watch?v=U8dcFhF0Dlk
I say this as someone with no skin in the game, but I don’t think it’s necessary for AI to replace human creativity in music. Instead, it can be a tool for augmentation.
There has almost always been backlash against new technologies in music. Ultimately, musicians adopted those technologies and continued producing great works.
I recently dug some 20-year-old CDs out from storage that were full of riffs and chord progressions for songs I never got around to completing. I fed them to Suno as a sort of “what might these have sounded like if I’d finished them?” experiment. I found the results quite impressive and plan to publish them some time in the near future.
None of the tracks sound like marketable products to me, and I certainly don't think AI is going to generate the next Bohemian Rhapsody -- but it definitely could be a great thing for curing writer's block.
We are creating animated corpses out of our culture.
It's interesting that the “Classical Music” sample ended up with a style prompt of “Contemporary Christian”.
It's even more interesting that the “Contemporary Christian” music it generated was specifically Mormon music (specifically about the exile of Lehi and his family from Jerusalem into the wilderness, during which they reach the ocean, build a boat, and sail to the Americas), but in the style of contemporary Christian rock.
The majority of people, I think, don't want to listen to AI music or view AI art. Systems like this will produce muzak for ubers, cafes, phone queues, advertisements. Some artists will make hyperreal work by sampling it. But to listen to these "songs" willingly is not something I expect most people will choose to do.
I went and listened to a bunch of the examples and it feels like it's trained in one very contemporary style. E.g. I hear a vaguely buildup-drop pattern that sounds a lot like EDM in places I wouldn't expect it.
I really want to like ai music but none of them seem to be trained on reward models that reward "ambiance" or any sort of interesting sound design (which this paper obviously doesn't even concern), which might be reflective of the people training them not having niche music tastes
Why do you want to like it?
Because there is a shortage of songs I like. I loop the same songs for weeks until I can find a new song
apt reference:
https://electricliterature.com/wp-content/uploads/2017/11/Tr...
"The Petty and the Small; Are overcome with gall ; When Genius, having faltered, fails to fall. Klapaucius too, I ween, Will turn the deepest green To hear such flawless verse from Trurl's machine."
With this local audio model, I can finally make proper covers in different genres and change lyrics.
Previously: https://news.ycombinator.com/item?id=46641627
That example "Comedy" sounds like the most soulless synthetic example of "music" ever. Like the music analogue of a PR-speak article of a big corpo generated by an LLM. It's still technically impressive but it's 0/10, absolutely empty of soul.
All AI generated art feels somewhat masturbatory to me, but AI generated music is the most baffling of them all. It's like what if someone thought Muzak had too much soul.
BTW: https://github.com/ScryptHunter/ComfyUI-YuE2
I guess it could be neat to play 100 of these at the same tempo at once but otherwise I'm having trouble coming up a use case for this...
Numbers Stations
You know how they used heavy metal music to break the spirit of Manuel Noriega? I'm sure there's some DoD (or is it DoW now?) psychological operations people right now studying the prolonged effects of high volume mediocre AI slop music.
https://www.bbc.com/news/world-latin-america-40090809
Or I suppose they could just put "baby shark" on a loop.
> https://www.bbc.com/news/world-latin-america-40090809
The playlist included "the sound of jet planes all night" and "the screams of rabbits being slaughtered". I don't think they really needed AI for that, anything loud would probably do the trick after 51 days.
Id prefer a burning hell to AI gen music if I were ever to have the choice…
if it turns Steve Reich, Terry Riley, and Brian Eno old man violent i'm right there beside them with a sack of detuned doorknobs
The use case is control. It's essentially a controlnet for the actual music writing. So a musician could compose melody, harmony, and lyrics and generate many performances of the same song until they get one they like. A non-musician could similarly generate the symbolic input from scratch, and perhaps they like the composition but not the performance. This allows them to re-render and re-render until they like the performance.
> a musician could compose melody, harmony, and lyrics and generate many performances of the same song until they get one they like
Can't relate to this. Art doesn't come out as a left-brained problem solving exercise or identifying the right product for a market. It's about expression. You do it when you're feeling things.
I'm a composer and I can 100% get behind this workflow. I will often compose on one instrument (guitar or keys) and then throw the midi into a DAW and experiment with sounds.
This is that.
Why.
Why have a feature that lets you get what you want?
are you kidding
I'm not species-ist. If a digital sentient being has something to say and they can express it through music, I'm more than willing to listen.
But listening to these samples, it's obvious that, while technically proficient, they have no voice. Funnily enough, neither does most of the human produced music slop that existed before gen AI.
All it exposes is that most people have no taste. This is as good as shuffling through popular songs on Spotify.
Try some good (read as great) underground techno.
Mainstream is for suckers.
Properly mixed ofc.
I have, I do. This is me:
https://www.discogs.com/artist/9699799-Neurojacker
Am I as good as Claudio PRC, Donato Dozzy, Ken Ishii or Takkyu Ishino? Not even close. But wanting to make more is reward enough for me to keep trying.