Hacker News Reader

Comment by gnarlouse | Hacker News Reader

I see music as "the space of all possible 5-second clips at stereo 48kHz 24bit depth". If you think about it, that space contains the intro to Stairway to Heaven and Oops I Did It Again, and the end of either song. It contains every 5-second segment of O Fortuna, plus a previously unimagined O Fortuna Remix with MF Doom rapping the pledge of alliegance backwards. My point is, AI is getting good at searching music space for novel patterns, and that's entirely the point of music, not making a career out of being an alcoholic minstrel with a tour bus.

The audience will out the good patterns, and it's up to the musicians or AI companies to serve better patterns.

View on HN

> and that's entirely the point of music

That's very reductive. Music, like writing or painting, is a medium, not a thing with intrinsic purpose. It can be a means of communication, sharing human experience, conveying emotion, evoking feelings, expressing a story, ...

To many, the relationship between what was written in the sheet and how it was played by a live performer, day I say virtuoso, is the imporant thing. The human component is the important component.

This is entirely separate from pop, which is the junk food of music - cheap, filling, bad for your health.

I would argue that the human element in music is not an important contributor towards the enjoyment of music for exactly 100% of people.

Humans just biologically enjoy rhythmic sounds. We don't care who makes it, we don't care how it is performed live. Those things are just hijacking emotional memory in heightened moments, but that is completely separate from the general natural enjoyment of music.

AI music simply drives the same point home that listening to random music does when you are without a care for lyrics or artists (often, and everyone).

Your point is asinine. You love a hell of a lot of music besides what you attach to live performance. Everyone does. You do not care about the human component.

Unless you want to lie and claim you're some musical purist, this argument is shallow thinking and nothing more.

Exactly. Gorillaz was literally the cartoon characters in my mind for the entirety of my childhood, before I knew who Damon Albarn was.

I would counter-argue that you're accusing the other party of exactly the shallow thinking you're guilty of - even worse, you cannot concieve of a thing, so you confidently state thing does not exist and the other party is at best a liar. One only has to have a cursory knowledge of performances on, say, pianos, to know that who and how plays an established piece was always a big deal.

You are missing the point entirely. As I already said: live performances are a completely separate type of enjoyment than music enjoyment. It is dishonest to ignore all of the music you enjoy that has nothing to do with performance, and to cherry pick the less than 1% of music you have likes that also overlaps with what you have seen performed live. Please, this only takes a couple seconds of actual critical thought.

I'm not saying AI music is inherently good. I'm saying good music is good music.

My take is literalist, not reductive: Music (and any art product) is configuration. The media involved are the physical laws of nature that present the configuration.

> It can be a means of communication, sharing human experience, ...

1. You're construing the journey with the result. The act of creating is not typically the result product itself.

2. Any perceived relationship to the artist via product is virtual and parasocial (respectfully stated). You have no relationship with Bach or Shakespeare or Michelangelo--you have an appreciation for their accumulated works. You may have a fascination for their stories and life through literary works. You have no relationship with your favorite artist, you have their works in your media library.

> "the space of all possible 5-second clips at stereo 48kHz 24bit depth".

That space is mostly noise.

Try cat /dev/urandom > /dev/audio for an example of the kind of noise that this space contains.

(If you stick with it and manage live long enough, you'll eventually hear a few bars from Stairway to Heaven. If I also manage to live long enough, then perhaps we'll be able to chat about it in a few million years.)

> I see music as "the space of all possible 5-second clips at stereo 48kHz 24bit depth".

5.6MB? That's an astounding number of combinations. 1 followed by 1733933 zeroes.

> If you think about it, that space contains the intro to Stairway to Heaven and Oops I Did It Again, and the end of either song. It contains every 5-second segment of O Fortuna, plus a previously unimagined O Fortuna Remix with MF Doom rapping the pledge of alliegance backwards. My point is, AI is getting good at searching music space for novel patterns, and that's entirely the point of music, not making a career out of being an alcoholic minstrel with a tour bus.

AI is not searching that space for novel patterns for the most part, it is taking what it has heard before and coming up with things based on that. Which isn't a dig at AI, that's pretty much how humans do things too. I don't think today's AIs would be able to come up with something like Stairway to Heaven if Led Zeppelin and the music they inspired had never existed though.

Agree. Maybe novel is the wrong term given my framing. "Listenable?" The combinatorial space of music is hyperastronomical, effectively infinite. And most of it is probably noise.

AI isn't "searching" in the standard indexing sense. But if say, Suno, is doing stable diffusion on fourier transform heat maps, and there's a finite space of configurations... it is using a heuristic approach to pick an option from a well-defined (gargantuan) set of options.

I can’t tell if this is sincere or parody. It is like you set out to write the most HN take on music possible.

Looks like https://clackernews.com/ is leaking!

I love music, I make music. It is sincere.

Reply on news.ycombinator.com