Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google's advanced music generation model and two new AI experiments

deepmind.google

Post reply on HN
Print view View original

61–70 of 361 posts

Re: Google's advanced music generation model and two new AI experiments

#61

Post by stefanos82 »

After a lot of thinking, I have come to a conclusion that expresses me personally, so feel free to disagree if you like:

The human being was not created to be deprived of the right to express his feelings through fine arts. From the moment you deprive me of the right to express myself as I feel through external as well as internal stimuli and you want to replace me with an insensitive mechanism that neither feels nor has judgment like a human being, then everyone will become cynical and unsmiling and all they will lose their meaning.

Re: Google's advanced music generation model and two new AI experiments

#62

Post by mypastself »

vouaobrasil wrote: post #11

From the article: > Each participating artist has partnered with us and will have a hand in helping us test and learn to shape the future of AI in music. Let me correct that: each participating artist has been given a wad of dirty cash in return for their help in training replacements for mass music. Notice that I said "mass music" -- people will always be able to create by hand, but it's clear that the ultimate goal…

> traditional musicians… will not be able to compete and therefore will not be able to do what they love to survive.

Why would anyone owe the musicians a job just because they enjoy doing it? I don’t expect anyone to defend my love of coding when I’m replaced by a future iteration of an LLM.

> Music composed by humans could still be better but AI will compete simply because it is cheap and widely available

If it’s truly better, the market for manmade music will still exist. If bland/poor quality art is no longer produced by humans, what’s the loss?

> It changes the minds of human beings to think more like machines

Not sure how this works, or why it’s so bad.

Re: Google's advanced music generation model and two new AI experiments

#63

Post by siva7 »

nervousvarun wrote: post #26
gardenhedge wrote: post #6

Wow, AI really will democratize music creating. This is an awesome way for people to translate their ideas to music without needing to be privileged enough to get violin or guitar lessons. I see a future where we all make our own music and share it freely with family and friends. This means less listening to manufactured pop songs written by suits with the sole aim of making a profit. This will still happen of course…

What “privilege” is required to learn to play guitar? Was Jimmy Hendrix privileged? Robert Johnson?

Time and discipline in a busy life. For most people also money for lessons from bored teachers if you're not a one-in-a-million like Hendrix.

Re: Google's advanced music generation model and two new AI experiments

#64

Post by bambax »

halflings wrote: post #41
bambax wrote: post #36

Some examples are interesting (the sax, the string ensemble); the "realistic vocal choir" is anything but. I tried riffusion and sang 10 seconds of "the house of the rising sun" [0] It produced... this... https://www.riffusion.com/riffs/d0464fb4-a53a-4ad8-a765-0a64... which, by some measure, is impressive (perfect recognition of lyrics, for one), but is also... a musical abomination? So IDK. Let's see where this goes…

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?

Re: Google's advanced music generation model and two new AI experiments

#65

Post by vouaobrasil »

amarant wrote: post #49
vouaobrasil wrote: post #11

From the article: > Each participating artist has partnered with us and will have a hand in helping us test and learn to shape the future of AI in music. Let me correct that: each participating artist has been given a wad of dirty cash in return for their help in training replacements for mass music. Notice that I said "mass music" -- people will always be able to create by hand, but it's clear that the ultimate goal…

As a indie game developer, this could plausibly enable me to add decent sounding soundtrack to my games. I couldn't afford to pay a proper artist, and my musical skills are... Limited.. It _is_ just a tool, a new instrument essentially, one that is easier to play than most. People being angry that there is now a new instrument that is much easier to play than the one they spent years to learn come across as entitled…

There is NO SUCH THING as "just a tool". All tools have societal effects that are inevitable in different ways because we operate on basal instincts most of the time, or at least our capitalistic society has made us fairly deterministic.

It's not the same as a new instrument that is easier to play, because with an instrument, you have to determine entirely what comes out of it. With this "tool", you only determine part of the output, a skeleton, and a machine determines the basic musicality.

I 100% do not buy the accessibility argument. Of course, everyone plays that card. But the overall gross societal effects of AI are not worth it, even if it does help some people.

Imagine a new nuclear technology that cures all cancer, but simultaneously allows anyone to create an atomic bomb in their backyard. Would you think this worth it? We have to consider the costs and benefits, not just the benefits like most technophiles think.

You may create something with this tool for your game, but I certainly won't buy your game if AI has been used in its creation.

Re: Google's advanced music generation model and two new AI experiments

#66

Post by midenginedcoupe »

vouaobrasil wrote: post #11

From the article: > Each participating artist has partnered with us and will have a hand in helping us test and learn to shape the future of AI in music. Let me correct that: each participating artist has been given a wad of dirty cash in return for their help in training replacements for mass music. Notice that I said "mass music" -- people will always be able to create by hand, but it's clear that the ultimate goal…

Amen.

I'd only add that the sort of music they're aiming at is already pretty much at your end state. Muzak (and its slightly more advanced cousin Library Music) isn't about human expression, but about a background noise to convey a loose mood. God forbid we should just listen to what someone has to say - it always needs a distracting background interference to go with it.

But, like it or not, creating this drivel does put the food in the mouths of a great number of professional musicians. Even big-name recording studios aren't busy full time with wonderful new creative experiments, it's jingles and library music that keep their doors open. But even this work has been declining sharply for decades, and this new wave of tools will just continue that decline.

If anything, it's going to entrench the music profession for the privileged as there'll be an even smaller pot of work available and so only those that can afford the massive time investment and comfortable with the risk of being without work will survive.

And all this because the people building and paying for these tools are themselves musically illiterate - they don't know what makes good music, or why muzak is utterly devoid of the thing they profess to love and democratise.

Re: Google's advanced music generation model and two new AI experiments

#68

Post by soco »

vouaobrasil wrote: post #56
kkzz99 wrote: post #47

Earlier quoted context omitted.

Every tool that makes it easier to create art is good in my book. In years past you had to actually have the instruments and had to know how to work them. Now you can much easier learn to use simpler tools to get even better results, which just increases the amount of people that can actually realize their creativity.

But part of the whole point of listening to music is enjoying the expression of the people playing the instrument. Music is much more than just the final output of a machine, it's an expression of the soul. You may be able to create music more easily with this tool, but I guarantee it will be soulless, just like Google. The results certainly will not be better. They will be a hideous amalgam from a computer. Just lis…

The sheer amount of pop music and its pop-ularity contradicts you a bit here. Because one can play live AI generated music just as well, hop around on the beat and prompt the viewers to put their hands in the air - a live performance is (arguably much) more than the music. But I may be wrong, maybe the pop assembly line is not the sign of commoditization I seem to see, and won't gain that much from AI support.

Re: Google's advanced music generation model and two new AI experiments

#69

Post by vouaobrasil »

mypastself wrote: post #62
vouaobrasil wrote: post #11

From the article: > Each participating artist has partnered with us and will have a hand in helping us test and learn to shape the future of AI in music. Let me correct that: each participating artist has been given a wad of dirty cash in return for their help in training replacements for mass music. Notice that I said "mass music" -- people will always be able to create by hand, but it's clear that the ultimate goal…

> traditional musicians… will not be able to compete and therefore will not be able to do what they love to survive. Why would anyone owe the musicians a job just because they enjoy doing it? I don’t expect anyone to defend my love of coding when I’m replaced by a future iteration of an LLM. > Music composed by humans could still be better but AI will compete simply because it is cheap and widely available If it’s tr…

> If it’s truly better, the market for manmade music will still exist.

That is definitely not true. There are other things that determine what is sold besides quality: ability to mass produce, economies of scale, bullying tactics, etc.

> Not sure how this works, or why it’s so bad.

Because thinking like machines means that we are locked into the cycle of optimizing resource usage to expand and grow forever and not be stewards of the environment. Thinking like a machine means being a human cog in the consumerist machine of selfishness. Thinking like a machine means facilitating the endless growth of technology that furthers us from connecting genuinely with other people in a small community.

Re: Google's advanced music generation model and two new AI experiments

#70

Post by vouaobrasil »

midenginedcoupe wrote: post #66
vouaobrasil wrote: post #11

From the article: > Each participating artist has partnered with us and will have a hand in helping us test and learn to shape the future of AI in music. Let me correct that: each participating artist has been given a wad of dirty cash in return for their help in training replacements for mass music. Notice that I said "mass music" -- people will always be able to create by hand, but it's clear that the ultimate goal…

Amen. I'd only add that the sort of music they're aiming at is already pretty much at your end state. Muzak (and its slightly more advanced cousin Library Music) isn't about human expression, but about a background noise to convey a loose mood. God forbid we should just listen to what someone has to say - it always needs a distracting background interference to go with it. But, like it or not, creating this drivel do…

Well perhaps we should try and rethink society instead of trying to make its generation more efficient...
Post reply on HN