HACKER Q&A
📣 simonebrunozzi

AI-Generated Music?


I'd guess that we would be now ready for great, successful AI-generated music. I know nothing about it, of course - can someone more knowledgeable than me tell me what's going on, what's interesting, what's noteworthy in the year 2022?


  👤 songeater Accepted Answer ✓
I've been working with OpenAi's jukebox to make vocals / "singer-songwriter" type stuff rather than electronica. Can do a HN writeup on process if there is interest...

Youtube Playlist: https://www.youtube.com/watch?v=_lcCJzfXl50&list=PLaa32nLgVv...

SoundCloud: https://soundcloud.com/songshtr/sets/machinecroon

Github: https://songshtr.github.io

The main challenge for me so far has been post-processing. Jukebox produces very "scratchy" / lofi outputs... and I've been DIY'ing with Audacity. Thinking of getting a track professionally mastered to see just how far the boundary is...


👤 peterlk
Has anyone solved the data problem? I have taken many stabs at this over the last 10 years, and the primary problem is that music is _extremely_ subtle. Even something as seemingly simple as chord progressions are very difficult to annotate properly without deep training. And even then, experts can disagree.

The thing that's missing is the data. If we had midi transcriptions of 100k songs (abc notation could be fine too), we could probably get really interesting stuff, but most of what is available is lossy chord transcriptions and classical music (public domain). So if you want to automatically create something that sounds like mozart, you're in luck!

But this isn't really satisfying to me. For generative music, we're still largely stuck with encoding musical rules in code rather than feeding data to a transformer. To me, the former feels much less like AI than the latter. The data is all locked up behind an impossible quagmire of copyright.

But if I were a sheet music publishing company, I would be seriously considering the future of music creation with AI given my broad access to notated music & metadata (is this an original score, or a grade 1 simplification?). But again, music copyright is a pretty complex contraption.


👤 sillysaurusx

👤 atomack
David Cope's been making algorithmic music for a while - not current but worth knowing about: http://artsites.ucsc.edu/faculty/cope/

edit: There's also this on ambient endless generative music that I think was an HN submission a couple of years ago https://generative.fm/


👤 simple10
AIVA has surprisingly good examples on their YouTube playlist that sound like real (human) compositions to my untrained ear.

https://www.youtube.com/playlist?list=PLv7BOfa4CxsHAMHQj0ScP...


👤 elpres

👤 adamnemecek
I’m working on something in this space

Launching soon

https://ngrid.io

Join the discord https://discord.gg/a5ttYuG


👤 winternett
One thing that's been pretty interesting is AI vocal splitting, which has revolutionized the art of remixing music that couldn't be fully remixed before.

I used it to remix one of my favorite songs ever by Biz Markie just before he passed last year. It ended up being the last remix made of his music before his passing, and I sent it to his manager just before, so I'd like to believe he heard it.

It still has a ways to go (as audio quality can be spotty and incomplete), but these online services can often separate more than just vocals now, they can cut individual instruments out of music, and even create pretty good instrumentals. I have been able to remove uncleared vocals from fully mixed tunes that I've made so they can be released as well. I never thought it would have been possible 20 years ago when I started music. As for AI generated music, I think it will be a travesty to de-value or remove humans from the music/art making process entirely, it will always likely be something derivative of human work in essence anyway, but I don't think it will ever match the depth and soul of human-generated music to people who truly know and love music, some things just can't be emulated.

Here's the remix video - https://www.youtube.com/watch?v=xL14JH5f-qM


👤 laserdancepony
Jean-Michelle Jarre released an app called "EōN" in 2019 that may be fitting to your description.

https://apps.apple.com/us/app/e%C5%8Dn-by-jean-michel-jarre/...

Article from BBC about the app: https://www.bbc.com/news/entertainment-arts-50335897


👤 charneykaye
Hiya! I'm the founder of XJ music. It's taken me seven years to build this thing I dreamed of when I was young, producing electronic music on a pair of MPC-2000s trying to hack them into something more.

XJ is human created content fed into a "player" that fuses the composition process with the storage and distribution. It's an algorithmic medium. The music is never "done" and what you hear in the app is playing live in real time from the software we run in the cloud.

What do you think??

Free on iOS and Android.

https://xj.io/download


👤 naillo

👤 shannifin
I'm actively working on a startup focusing on this (while doing startup school this month). For now I'm focusing on the melody generation, as to me that seems to be the area most lacking in AI-generated music, yet what I find the most interesting to work with and explore as a composer.

That said, I don't have anything to show yet... Only got a landing page at https://tunesage.com at the moment... but here's an ancient prototype writing very generic melodies just to prove I'm trying... https://www.youtube.com/watch?v=8cwEaiwEZfU

Regardless of my personal eventual success or failure in the endeavour, the field does seem to be ripe for innovation! Nothing out there at the moment (that I've seen at least) seems to satisfy my current desires as a composer.


👤 Rochus
Here is a recent, interesting survey paper: https://hal.archives-ouvertes.fr/hal-01660772

From my perspective as a trained, practicing musician MuseNet produces quite credible results: https://openai.com/blog/musenet/. But we're not yet there; AI generated music is still recognizable as such. Another, earlier project with very good results was https://www.microsoft.com/en-us/research/publication/automat... unfortunately the website is no longer active.



👤 balthigor
Entirely AI generated music. Lyrics generated by GPT variants, album art by diffusion models, music composed and performed by Jukebox (customized) - http://soundcloud.com/baltigor


👤 chris_st
Not entirely AI-generated, but one of the two folks on this project used an AI music generator (called "Jukebox", IIRC, but it's not on the page) to generate snippets based on the other's musical ideas, and then the two collaborators worked with those to build songs.

Some nice results, here's the "About" page that describes their process:

https://ooo.ghostbows.ooo/about/

----

Edit: Here's where Robin Sloan mentions Jukebox:

https://www.robinsloan.com/lab/notes-on-a-genre/


👤 kristiandupont
There is this: https://www.aiva.ai/

I've toyed around with it a bit. It's impressive for sure, but I am not sure I think of it as anything other than a curiosity.


👤 newaccount74
It's an older project, but I love Infinite Gnossienes:

https://gnossiennes.mousereeve.com/

It generates an endless version of the famous minimalist piano piece.


👤 riidom
https://www.youtube.com/c/DADABOTS_official/about Death Metal, the only music robots can love.

👤 nobodywasishere
Here's 10 hours of prorcedurally generated djent

https://m.youtube.com/watch?v=52SRWJ2pMXI


👤 Ashbt
Made a bunch a while ago using Jukebox and cleaned it up a bit using some izotope plugins: https://m.soundcloud.com/R05c0

It's really just a matter of time before they figure out how to make the music a bit more "cohesive."

Right now it's kinda like a stream of consciousness, vs. being verse, chorus, verse, chorus.

My prediction: in under 5 years, most of us will have a favorite AI band.


👤 hansword
https://stevenwaterman.uk/musetree/ - open front end for MuseNet (by ClosedAI), choose your model characteristics, then right-click on a node to request new bits, choose which you like best, repeat.

👤 AriedK
https://www.ampermusic.com/ Amper music, owned by Shutterstock, has been around for a while. It was quite nice to play around with when it was still free, not sure if, and how it has evolved over the past years.

👤 impetus1
Funnily enough, there is something on quitrit(quantum computing) generated sound representation which does generate music.

https://github.com/scottoshiro2/IBM_QCHack_Challenge2021


👤 WithinReason
Here is one of my favourite YouTube channels, Two Set Violin (professional musicians) listening to an AI generated symphony:

https://www.youtube.com/watch?v=R69JYEfCSeI


👤 adzm
The first step is AI-assisted music, really. And that's a great and useful niche. Think auto accompaniment, backing tracks, etc. Tools for helping figure out transitions. Someone already pointed out magenta which is a great starting point.

👤 pavelegorkin

👤 smugma
UK government seems to be supporting it (AI qualifying for copyright):

https://news.ycombinator.com/item?id=31920524


👤 kwatsonafter
This is a song composed with AI in the style of The Fab.

https://www.youtube.com/watch?v=LSHZ_b05W7o


👤 rogierhofboer
https://www.musicdevelopments.com/

Technically not ‘AI’ though, more based on music theory ‘rules’.


👤 bj-rn
klingklangklong has a few nice AI/ML based projects, for example:

https://klingklangklong.com/project/meandering-river

https://klingklangklong.com/project/for-seasons


👤 spacec0wb0y

👤 a-dub
a few years ago i went to a talk by some people that were working on this kind of stuff within google brain. unfortunately i don't remember any names though.

edit: just had a look. it appears to now be the magenta project which is referenced elsewhere in this thread.


👤 suyash
Look up Google Magenta projects. We made a web app using Magenta.JS VAE model : https://devpost.com/software/dear-diary-ezrmgt app : https://deardiary.ai/

👤 alxmrs
I would love to see a LM generate sonic-pi scripts which generates music.