OT: Has anyone tried the opposite - ask AI to listen to music and determine the notes or chords being played? Or watch someone playing an instrument and give a textual output of what notes/chords they are playing.
I would love this! There's a song I like by a band that broke up in 2013 and I am transcribing it by watching a live performance they did and trying my best but realizing I'm trying to take a mandolin/guitar and put it to acoustic. Even just being able to do a similar rendition would be nice by telling the AI "hey, do a twist on this and give me the chords/tabs".
I'm very interested in this too. We're beginning to see models that can deeply understand a single image, or an audio recording of human speech, but I haven't seen any models that can deeply understand music. I would love to see an AI system that can iteratively explore musical ideas the same way Claude Code can iterate on code.
Reminds me of an example in a similar direction, where AI was used for audio processing to filter out everything except a person's voice. If I remember right, it was able to focus on different people in a crowded room. It might have been also for music, to pick out an instrument and listen to it, filtering out the rest of the band.
There are several tools that do this already - AnthemScore, Spleeter, CREPE, and even Google's AudioLM can transcribe music to MIDI with varying accuracy depending on instrument complexity and audio quality.
I use https://moises.ai/ multiple times a week for practicing / figuring out chords being played. For the notes (say in a guitar riff), I dont know if such a thing exists
I did this for my graduate capstone (https://www.deepjams.com/)
We extracted chord progressions from existing music you would upload and then riffed based on those chords. there are open source libraries for this.
Is it more or less dystopia than an army of musicians trying to eke a living out of creating stuff like this (https://www.chosic.com/free-music/presentation/) to go behind your company's PowerPoint about how its Q3 woodchip sales didn't quite exceed expectations?
Maybe the fundamental issue is that this shouldn't compete with a human picking up a guitar and having fun with it, and the only reason it does is because we keep tying questions like "survival" to whether someone can make woodchip earnings reports less boring to read instead of trying some other way to be a community?
Just wait until the rush of commenters that insist you’re wrong about being offended that they want to automate and commodify every aspect of your life.
The thing derpressing to me is all the energy spent making stuff I can enjoy doing or building weapons while the tech industry is still unable to solve real problems and build me a generic cleaning robot that will make my laundry, clean up my bathroom and toilets, wash the windows. All we have are shitty and useless autonomous vaccuum cleaners that get stuck at every other piece of furniture with legs or the first sock on the floor.
A lot of criticism of AI in music seems to be around the lack of originality and it being generic slop. But unfortunately that’s true for pretty much the entire music industry. Handful of people write songs for all the artists, “artists” don’t create the songs as much as they perform them, most of the music isn’t created but rather sampled or is “inspired” by other music and pretty much most of the artists sound like other artists.
Yes there are smaller creators who are trying to make something net new, but unfortunately 99.9% of the small artists are also derivative and lack originality.
I see AI music as just continuation of the sad state of the industry at the moment. Hopefully it accelerates the demise of the industry as we know it and restarts the cycle of creation.
Well, so much for culture. Every use case where you can plausibly use AI generated music removes one more method that provided an avenue to a reasonable career making music.
If you consider that Entertainment gives the viewer what they want, and art intends to challenge, none of what's created here is "art". It doesn't push boundaries, create new genres, or satisfy an uncomfortable curiosity.
The tech here is fantastic. I love that such things are possible now and they're an exciting frontier in creation.
It's very dystopian to feel that the robots are making generic human-music with indescribably lifeless properties. I'm not an artist, so I don't feel personally attacked. Much like image gen, this seems to be aimed at replacing the bare-minimum artist (visual or auditory) with a "fill in the blanks" entertainment piece.
I could be wrong, but I think the use case here is mainly for non-artists in domains where the music is not particularly important.
For example, a podcaster/youtuber may want a short intro track. An entertainer or a marketer may want some generic or silly background music.
Does it have a use case for a producer/musician? Maybe. It might give them ideas for chord progressions, melodies, etc. But real music does that too, and much more effectively.
Well, they've done it boys, they've made creative fulfillment obsolete. They've DISRUPTED the concept of going to big music festivals, and small cozy shows. Just plug your ear holes with the AI slop bucket's pure beeps and boops and never have to worry again about paying artists for music. You can pay a techbro instead.
This is like the dotcom era of where every idiotic idea that ended with, "but on the internet", would get a pile of cash thrown at it. We are officially at the beginning of the end. It's only going to get dumber from here.
We imagined a utopian future where robots did our menial work so we were free to be creative. Instead we got a dystopian future where we do more and more menial work so our robots can poorly emulate creativity. It's not too late to turn it around, but that requires recognizing the humanity of 99.9% of people, and the 0.1% who own everything would rather create their own synthetic (subservient) humans than recognize the basic rights of the ones that already exist (and can make fun of them on Twitter).
Behold, we are one step closer to a world where "AI will do the mundane tasks, and the we all are going to engage in creative hobbies such as music creation"... Oh... wait...
This is an incredible achievement, but as a musician, I wish this would go die in a fire.
I'm a bedroom hobby musician with no dreams of ever making it big, but even so, I'm looking at the hours I'm spending trying to improve my skills and thinking what's the point, really, when I could just type in 'heavy metal guitar solo at 160bpm, A minor' and get something much much better?
I know there is value in creating art for art's sake. I've always been up against a sea of internet musicians, even when I started back in 2000. But there's just something about this that's much more depressing, when it's not even other people competing with me, but a machine which hasn't had to invest years of its life in practice to beat me.
What struck me more than the output is the absolute marketing talk of the prompting, so not only is the output kind of creepy but in order to get the best results you have to translate your intention into the worst corpo-speak imaginable.
72 comments
[ 9.4 ms ] story [ 61.5 ms ] threadMaybe the fundamental issue is that this shouldn't compete with a human picking up a guitar and having fun with it, and the only reason it does is because we keep tying questions like "survival" to whether someone can make woodchip earnings reports less boring to read instead of trying some other way to be a community?
Yes there are smaller creators who are trying to make something net new, but unfortunately 99.9% of the small artists are also derivative and lack originality.
I see AI music as just continuation of the sad state of the industry at the moment. Hopefully it accelerates the demise of the industry as we know it and restarts the cycle of creation.
The vocals are definitely not that.
The tech here is fantastic. I love that such things are possible now and they're an exciting frontier in creation.
It's very dystopian to feel that the robots are making generic human-music with indescribably lifeless properties. I'm not an artist, so I don't feel personally attacked. Much like image gen, this seems to be aimed at replacing the bare-minimum artist (visual or auditory) with a "fill in the blanks" entertainment piece.
For example, a podcaster/youtuber may want a short intro track. An entertainer or a marketer may want some generic or silly background music.
Does it have a use case for a producer/musician? Maybe. It might give them ideas for chord progressions, melodies, etc. But real music does that too, and much more effectively.
This is like the dotcom era of where every idiotic idea that ended with, "but on the internet", would get a pile of cash thrown at it. We are officially at the beginning of the end. It's only going to get dumber from here.
"That is not allowed by our terms of service"
I think the rebellious nature of art inherently has boundaries these people won't cross.
It's amazing that the songs sound pretty natural
I'm a bedroom hobby musician with no dreams of ever making it big, but even so, I'm looking at the hours I'm spending trying to improve my skills and thinking what's the point, really, when I could just type in 'heavy metal guitar solo at 160bpm, A minor' and get something much much better?
I know there is value in creating art for art's sake. I've always been up against a sea of internet musicians, even when I started back in 2000. But there's just something about this that's much more depressing, when it's not even other people competing with me, but a machine which hasn't had to invest years of its life in practice to beat me.
I am less interested in the "one-shot" approach here with text-to-prompt. I see seamless transitions but that seems like an afterthought.