Because musicians haven't been hurt badly enough already by big tech
Google's advanced music generation model and two new AI experiments
351–360 of 361 posts
Re: Google's advanced music generation model and two new AI experiments
#352Michael Jackson was an excellent musician, who just happened to not play any instruments. He'd hum lines, beatbox rhythms to his band who'd then take his ideas and transform his ideas into instrumental music. Most people don't have access to the quality musicians MJ had, but there's a lot of creative people that lack the skills to create their artistic vision. Tech like this could be the unleashing of an age of new g…
The next michael jackson will be an ai generated 3d model placed in ai generated videos with an ai generated voice making ai generated music, owned wholly by some big entity like zenimax or Disney. AIs like that will swamp the internet with so much stuff that it will become impossible to find authentic content, much like it's almost impossible to find a blog that isn't seo optimized and basically a business these days, except 10x worse since the ai output will be so large. This will be the end of non-live music and of celebrity musicians
(until: smartphone companies generate personalized musicians and directors and writers for you, and then even Disney gets sunk)
Re: Google's advanced music generation model and two new AI experiments
#353Earlier quoted context omitted.
I would argue that this has already happened and it's only a matter of degree now. The widespread availability of DAWs, streaming services, and music distribution services has greatly enabled people to create and publish their music online. Not to mention services like Splice which enable the purchase and curation of ready-made musical loops. It's never been easier to make a record, render it, and distribute it. Of c…
Yeah the democratization era is already well underway: arguably over, even, in that it’s no longer possible to become a “big star” from throwing together some shit in fruity loops and hitting it big. Every once in a while, a bedroom musician with a little talent will break through, but more and more music is either relegated to “good taste but unknown” or “weird conglomerate of ghostwritten tunes operated by a Hollyw…
Why do you think it's great?
Re: Google's advanced music generation model and two new AI experiments
#354Earlier quoted context omitted.
While the mixing process is certainly a frustrating one, it’s also the part I’m most addicted to. I mix something almost every day, whether it’s my own work or other multitracks “acquired” from the internet. It’s so much fun working backwards on a song I’m familiar with and going “ok, how the hell did they make the kick sound like that?!” and figuring it out :)
Where can one find multitracks on the internet?
Cambridge Music Technology has a great list of multi tracks for learning the art of mixing. The submissions cover many genres.
Re: Google's advanced music generation model and two new AI experiments
#355Earlier quoted context omitted.
Not that much difference. You used to have to gather 4 musicians to form a band, pay for recording studio, record an lp, and tour to get your music out. This took years to get that sort of creativity out. Now you make a song on your laptop with some existing samples, and release it to SoundCloud. So it takes much less technical and theoretical knowledge to create a song than 70 years ago. This is just an extension of…
No. You no longer have to gather other humans. But electronic music production requires a wild amount of technical knowledge from a single individual, whereas being in a band only requires you to know how to play one instrument. To do electronic music production decently, you need to understand music theory, the actual physics of waves and how to manipulate that (synthesis), how to even process samples (you can't jus…
We are actually kind of agreeing.
Let me change my words a little bit then.
The barrier to entry was significantly lowered through electronics. I was making music on Fruity loops when I was 12 years old. It was shit, but I had everything I needed in a single program. Becoming a producer however is extremely difficult.
The same will happen with AI. The barrier of entry to making something nice will be even lower. But mastering the AI tools, and really using it and combining it with musical knowledge will be extremely hard.
And those people will rise to the top of creating new musical genres, with the inclusion of AI.
Re: Google's advanced music generation model and two new AI experiments
#356To musically trained people, this is so weird. From the page:
"Imagine singing a melody to create a horn line, transforming chords from a MIDI keyboard into a realistic vocal choir, or adding an instrumental accompaniment to a vocal track."
This is what composers do! It's not hard, it requires a journeyman level of skill and training. I guess it will change how non-musicians create really bad music though?
I'm not saying it's not a game changer - the crappy library music world is running on borrowed time. But positioning this as revolutionizing music production is just bizarre. It's not solving the problems that one faces trying to write good music - it's just making the simplest steps faster. And only significantly faster for the untrained.
Re: Google's advanced music generation model and two new AI experiments
#357To musically trained people, this is so weird. From the page: "Imagine singing a melody to create a horn line, transforming chords from a MIDI keyboard into a realistic vocal choir, or adding an instrumental accompaniment to a vocal track." This is what composers do! It's not hard, it requires a journeyman level of skill and training. I guess it will change how non-musicians create really bad music though? I'm not sa…
I'm a bit confused why so many models seem to focus on text-to-audio.. what use is an mp3 in a composition? I just need the notation. One could take an avant garde approach and find ways to make use of the audio outputs, but that doesn't necessarily serve "normal" music.
Re: Google's advanced music generation model and two new AI experiments
#358To musically trained people, this is so weird. From the page: "Imagine singing a melody to create a horn line, transforming chords from a MIDI keyboard into a realistic vocal choir, or adding an instrumental accompaniment to a vocal track." This is what composers do! It's not hard, it requires a journeyman level of skill and training. I guess it will change how non-musicians create really bad music though? I'm not sa…
I'm a bit confused why so many models seem to focus on text-to-audio.. what use is an mp3 in a composition? I just need the notation. One could take an avant garde approach and find ways to make use of the audio outputs, but that doesn't necessarily serve "normal" music.
One can only assume they have decided their market is people who want to type some stuff in and then pat themselves on the back for "composing". It's all accompanied by ballyhoo about "democratizing music production".... which happened in the late nineties when you could run a DAW on a home machine.
This has nothing more to do with democratizing music production then audio to text has to do with democratizing writing a novel!
Re: Google's advanced music generation model and two new AI experiments
#359Earlier quoted context omitted.
I think Max Tegmark said there's 3 main camps of people talking about AI: 1. The people who think AI won't really work (at least in some fundamental way - it might make some stochastic melody but it won't have any creative spark to it). 2. The people who think AI will be able to do everything better than humans (in every possible way, at least as far as a mere human can tell) and that this will be great. 3. The peopl…
4. The people who think that AI will be able to do everything worse than humans, but at an acceptable enough level for executives, and it's far cheaper than even outsourcing work to other countries. The trajectory of the last 10 years has been that the quality of the human experience will go down so the people at the top can make more money, and AI fits perfectly into this narrative.
People used to say that this will never be accepted, especially in high stakes areas like driving a car, but then we saw how it works in the real world.
Re: Google's advanced music generation model and two new AI experiments
#360Earlier quoted context omitted.
Growing pains. Remember how this all started (Google Magenta, circa 2019): https://www.youtube.com/watch?v=ZRnbbtqxBEc
It seems most (or all?) AI music projects aim at producing audio? Why not output MIDI instead, and let the artist manipulate that? I for one would gladly buy a VST instrument that would generate a constant stream of MIDI ideas, chordings, voicings, variations, from a few seconds of singing or whistling or clapping. It seems the technology is here to build it, yet (AFAIK) it doesn't exist. Why?
> Why not output MIDI instead, and let the artist manipulate that?
This is not true at all, it all started with generating MIDI, and even the very link I shared is of a system that generates MIDI.