Related: <a href="https://blog.youtube/inside-youtube/ai-and-music-experiment/" rel="nofollow noreferrer">https://blog.youtube/inside-youtube/ai-and-music-experiment/</a><p>(via <a href="https://news.ycombinator.com/item?id=38298670">https://news.ycombinator.com/item?id=38298670</a>, but we merged the comments hither)
Michael 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 genres and musical creativity. I'm optimistic.
The older I get, the more I appreciate craftsmanship. Applies to nearly everything. Lately, I’ve been sitting in a chair and listening to records - not doing anything else. There’s a connection to myself, the music and the artist that’s different than when I’m listening on Spotify with my AirPods on.<p>To each their own (I mean that) but I do wonder sometimes…At some point, what’s the point?
This article is such a stark contrast from the GraphCast AI weather model they released [1]. The latter focused so much on the tech and what it does do, whereas this article buries the tech in a bunch of fluff about what it could do.<p>Can you imagine if the GraphCast article were like that? "Revolutionizing the future of weather".<p>I'm sure this music thing is cool but I have no idea what was built.<p>[1]: <a href="https://deepmind.google/discover/blog/graphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting/" rel="nofollow noreferrer">https://deepmind.google/discover/blog/graphcast-ai-model-for...</a>
In "Brave New World," all the music is made by machines. We are approaching that dystopia. For me, much of the pleasure in listening to music lies in appreciating the virtuosity of the musicians. It's like watching a great tennis player or a great dancer. You admire their skill and grace and technical mastery. For me, listening to music made by machines would be like watching robots playing tennis, or robots dancing. Pointless!
Wonder where the human fits in this new landscape. Composer? no. Performer? no. Music artist is more like a conductor now. Thankfully audience doesn't care because they need to worship a human in the end, so she 'll be receiving all the money anyway.<p>The problem with these new BigTech companies is that they are accustomed to subjugating everyone. They deal with users and content, not with artists and creations. This is starkly different from the past. Yamaha and other synth makers would go out of their way to satisfy capricious artists, now it's the other way around, the tech dictates to artists what is possible, including the limits of what they are allowed to make/say and sell.
Not sure I agree with the negative opinions in this thread. Surely, having an additional option to listen to the song you want to is a good thing? It's not like the choice to listen to human artists is taken away. Furthermore, a lot more people can now express themselves through music that were previously unable to.
It's really interesting for me to see AI voice cloning taking off. A few years ago I worked on a paper called HuBERT (<a href="https://arxiv.org/abs/2106.07447" rel="nofollow noreferrer">https://arxiv.org/abs/2106.07447</a>) which has since really taken off for doing this kind of stuff (it's the speech representation that SoViTs uses, for example). At the time the main research focus was low-resource ASR; it was sort of a fluke that it works so well for voice conversion. It makes me feel confident that open-source ML will drive more value over the next five years than closed-source ML, despite all the hype around big tech, simply because random people building weird, unexpected stuff on top of foundation models will make things that no one could have predicted.
Replace "Transforming" with "Disrupting" and the story feels more real. Frankly, I love music. That is why I'd prefer if Google did not mingle in this bussiness. Creatives already have a hard time making any sort of living from their talent, and AI generated "elevator music" will not make that situation any better.
Some examples are interesting (the sax, the string ensemble); the "realistic vocal choir" is anything but.<p>I tried riffusion and sang 10 seconds of "the house of the rising sun" [0]<p>It produced... this...<p><a href="https://www.riffusion.com/riffs/d0464fb4-a53a-4ad8-a765-0a64650f36f9" rel="nofollow noreferrer">https://www.riffusion.com/riffs/d0464fb4-a53a-4ad8-a765-0a64...</a><p>which, by some measure, is impressive (perfect recognition of lyrics, for one), but is also... a musical abomination?<p>So IDK. Let's see where this goes.<p>[0] <a href="https://www.youtube.com/watch?v=4-43lLKaqBQ">https://www.youtube.com/watch?v=4-43lLKaqBQ</a>
From the article:<p>> 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.<p>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 is to create a system whereby people pay for AI generated music (either with money or with ad views) and no longer need human-created music for commercial purposes.<p>> Our music AI experiments have been designed in line with YouTube’s AI principles, which aim to enable creative expression while protecting music artists and the integrity of their work.<p>This is hardly a protection. It only protects existing, secure artists. It does not protect new artists from being outcompeted by AI. (Music composed by humans could still be better but AI will compete simply because it is cheap and widely available.)<p>This article written by Google is a testament to the true nature of Google, which is the nature of being a leech and parasite. OF COURSE, at first, this tool MIGHT be interesting and provide new ways of creating music, but we have to remember that it will mainly function in the context of YouTube, which is about superficial consumption and advertising. The addictiveness and widespread usage of YouTube is becoming a societal problem, and this technology will standardize AI music into becoming mainstream.<p>Many people think that AI is just a tool to aid in music creation. It is for now, but what it actually does is three things:<p>(1) It creates a new standard where music is no longer about human expression but about creating a basic layer of entertainment to go along with advertising.<p>(2) It changes the minds of human beings to think more like machines.<p>(3) At some point, it will be so advanced that it will hardly need anyone to create it, and Google and other big tech companies can use it to be an alternative to traditional musicians, who will not be able to compete and therefore will not be able to do what they love to survive.<p>Remember, technology at first is an optional curiosity but successful technologies ALWAYS become entrenched in society and modify it significantly, and in this case, for the worse.<p>Shame on Google for this and I truly despise them.
What we do as humans is to push against a wall of mist of a unlimited search space.<p>Doesn't matter if it is music, drawing or scientific discoveries.<p>What an Artist does is pushing the boundaries.<p>Having tools like ths allows much more people to push and not 'just' the artists who mastered certain skills.<p>I can write very good software, do i care if my family suddenly can also write softwre thanks to ai? Honestly no. My goal was never to write code but always to create something which adds value to my life or to others.<p>I would not have thought that we will achieve this level so fast but yes writing code and using certain tools will be not necessary in the near future.<p>But hopefully we will have more time to learn all of those skills for the sake of having fun learning them.<p>I have a very good sense of what music sounds good to me and what art looks good to me. Should i be not allowed to create my own art even if its through tools like SD, Midjourney and co?<p>don't be that type of person
The tech is amazing, if it wasn't the comment section would be about how it's not good and not as melancholic as it is here. The melancholy is a testament to the capability.
To musically trained people, this is so weird. From the page:<p>"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."<p>This is what composers do! It's not hard, it requires a journeyman level of skill and training. I guess it will change how <i>non-musicians</i> create really bad music though?<p>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.
There is an extremely weird juxtaposition between the idea of watermarking the outputs so AI-generated samples can be detected if they make up even <i>part</i> of a song (which is arguably considerable as fair use of even <i>copyrighted</i> material if only a single bar is sampled), and the statement:<p>> Within the experiment, a limited set of creators will be able to use Dream Track for producing a unique soundtrack with the AI-generated voice and musical style of artists including Alec Benjamin, Charlie Puth, Charli XCX, Demi Lovato, John Legend, Sia, T-Pain, Troye Sivan, and Papoose*.<p>So... we're watching you and we'll know if you so much as <i>sample one bar</i> to create something of your own. But our preferred creators are having so much fun using AI exclusively to make the entire composition, <i>directly ripping off the personal style</i> of these artists who said it's OK!<p>I also found it hilarious that they claim the idea of embedding a watermark in the audio by adding the image to the 2d spectrogram is a "novel method ... unlike anything that exists today, especially in the context of audio." Considering Aphex Twin did it in 1999...<p>* Apparently it was removed from the game, and his Wikipedia doesn't even mention it, but, yes, Papoose, of the "You just mad cuz I'm stylin' on ya!" track from GTA IV.
Hahahaha poor naive people that accepted to "collaborate" with Google, Google has taught us how their seemingly benign acts always end up as they started, right?<p>Nothing more trustworthy than a multi billion corporation trying to destroy the internet as we know it to get more billions out of ads convinces a bunch of washed up artists that their new project is benign and even are so good that they use the hashtag "Social Responsibility".<p>This is very totally not just a plan to try to get others on board, wonder what's the catch here? try to do a FOMO where only the first few get to get rights for their voice being used? playing a very long con getting voice exclusivity rights? or what exactly?<p>One thing I know is that whatever Google says is a lie.
The slander in this thread against Top 40 is depressing if not unexpected. The music sounds good, what's the problem? Musical elitism is one of the oldest of the modern tropes, and deserves no attention paid to it.
I believe it will still be worth it to become musically trained; otherwise, you’d just be adding to the pile of worthless garbage on redundant storage somewhere. My only fear is how to preserve the idea with kids of how awesome and enlightening it is to learn to play a musical instrument, be able to master it, even hacking it, and maybe improving it. That’s why over time I’m appreciating more and more analog music producers like Vulfpeck, and their success might be a lesson to learn from, even though I don’t expect many to know them in this high-tech context.
> SynthID embeds a watermark into AI-generated audio content that’s inaudible to the human ear and doesn’t compromise the listening experience. It does this by converting the audio wave into a two-dimensional visualization that shows how the spectrum of frequencies in a sound evolves over time. This novel method is unlike anything that exists today, especially in the context of audio.<p>Is this a unique or boolean watermark (==ai_generated)? Sounds like it would be amazing to have for regular YouTube uploads to prove authorship and deal with reuploads/content theft.
The watermarking is worrying. I can imagine many different vendors will start adding something like that to their products - at some point it has to interfere with the signal, no?
Google is pursuing the absolute worst of what AI music could be. It's really disappointing. Instead of elevating human expression they went straight for generating generic mass-market music to back Shorts videos. Zero influence from human creators that can actually make things great. Models can reproduce style, but they can't reproduce taste and emotion.
Audio quality is not great yet, but the harmony is ok. Also, there's a video where a beatbox is transformed into a drum loop - the result doesn't match the source perfectly, AI has added some crash (?) and extra shakers.<p>Also, as I understand you must be able to sing properly to use this tool, so it is only for talented kids, and ordinary kids still have to use a keyboard?
AI will remove the traditional 'Creator'. Now, it's a direct line from platform to consumer, with the AI as the new creator. No royalties, no middlemen, just pure, content delivery... Creators better wake up, or have something to license to the platform...A gait, a singing style, a voice tone, a guitar or piano style...
The moment quality in art becomes commodified, what is popular will select on something other than quality. Look what happened to visual art in the earliest 20th century. Any art academy graduate could give you a beautiful painting, but nobody remembers any artists from the 20th century who simply made "beautiful paintings"
I think this is fanastic. I just read an article where Vince Clark talks about using 1 finger to program his synthesizer and he's written some of the best pop songs.<p>Imagine the talent AI assisted music programs will unlock. I think it will be a new instrument people learn to play.
> 9 artists ... who have chosen to collaborate in this experiment and work with us to shape the future of AI in music<p>I really hope it's not google that drives the future of AI in music.
I never want to hear this stuff but you know they're going to make it available for every content junkie publishing hype beast low effort mouth open consumer product placement having youtuber so it's going to be forced into your ear holes at some point. I hate google so much.
Yet another tool from Google that's closed. No model, no code, no real details. You can't even run the model today.<p>Google continues to live in the darkest timeline when it comes to ai and ml.
I hate this much less than I thought I would. It seems great for prototyping. My kids would love it. If you're trying to make a finished track, you're going to want to jump some of the rails the AI puts in place, and at the end of the day, the map is not the territory, so someone will need to be able to make the exact sounds and music that you want, unless you are willing to go with whatever the AI puts out there. I anticipate this is more than just a toy, but still confined to something like a genre of music, albeit one that could eat most of the industry (see rock vs classical).
I might have a different take, but the introduction of ChatGPT last year actually prompted me to dive head first into learning an instrument aided also by a fantastic live performance by Yo-Yo Ma.<p>In one interview I watched, Yo-Yo said he strived to play pieces perfect until his 20s when in the middle of a concert he realized he was playing perfectly and it was boring. From that moment he decided to dedicate his musical performances to making a <i>connection</i> with the audience. His concert was incredible to experience a room of 1000 people all having their seats <i>shaken</i> by a human being bowing a string without any electronics involved. People were getting drawn in, then jumping back, and getting transported outside the physical. Hard to describe.<p>And that was when i realized that no matter how perfect a computer plays, humans will always want to hear other humans play. We have had player pianos for over a century and not to mention a long history of sculptures that play music and even now you can get a virtually unlimited amount of extremely high quality near perfect replication recording of the best performances out there.<p>But still, I just went to buy tickets to a classical music event on a monday night to a solo concert by a not-famous player that each seat cost around $100 and they were sold out.<p>For me personally, learning an instrument has been a tremendous experience. Even the interaction with my teacher gives me weekly human contact that has become scarce in the post-covid world filled with remote work.<p>Spending months of practice just to play a song poorly is actually an incredibly freeing and humbling experience. I had my recital with a bunch of people watching me, where I messed up, I played out of tune, but it was an incredible flood of human connection that i never really had prior.<p>I think worrying about AI taking over music is overblown. Unless the music is going to be generated to manipulate people somehow, but the reality is that in 100 years people will still crave live performances. Just like for the past 100 years, paradoxically, the access and lowered barriers of entry, has actually increased accessed to classical and live performances. People complaining that people today don’t like “music with passion” anymore are living in some fairytale land where they imagine every single person 200 years ago was skipping down to the opera house. The reality was that 200 years ago access to music was incredibly limited and most people only experienced it in church context. If that (rural churches had a lot less going on musically).<p>People will always crave to see other humans’ accomplishments. Music, paintings, etc are not just an end product, but the more you understand the difficulties and journey of the person producing it, the deeper your appreciation becomes.
After a lot of thinking, I have come to a conclusion that expresses me personally, so feel free to disagree if you like:<p>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.
To quote from 1984:<p>> The tune had been haunting London for weeks past. It was one of countless similar songs published for the benefit of the proles by a sub-section of the Music Department. The words of these songs were composed without any human intervention whatever on an instrument known as a versificator. But the woman sang so tunefully as to turn the dreadful rubbish into an almost pleasant sound. He could hear the woman singing and the scrape of her shoes on the flagstones, and the cries of the children in the street, and somewhere in the distance, a faint roar of traffic, and yet the room seemed curiously silent, thanks to the absence of a telescreen.<p>Part of me worries that I am just gatekeeping.<p>However to me, the point of music was that its meant to put across something personal, human and emotional. Even when its a cover. Whether its a protest song, or shouting about taking someone to a gay bar. To automate that away, what does that do to us culturally?<p>So much of our identity comes from shared art. Emos liked emo music (you know with the shouty backing singer), goths went for death metal or the cure. Indie twats waxed lyrical about how heroine was great for the libertines.<p>All of that shared (mostly bullshit) memory is going to disappear, so what will bring us together, unite groups of kids with a shared like of x? Will it matter?<p>I worry that all the things I enjoy doing are being automated away. Perhaps this is what the weavers felt like when they were replaced with powered looms. Or cabinet makers felt when they saw steam powered power tools that sparked the arts and crafts movement[1]<p>I also frankly worry about how my kids are going to make money.<p>[1]<a href="https://en.wikipedia.org/wiki/Arts_and_Crafts_movement#Critique_of_industry" rel="nofollow noreferrer">https://en.wikipedia.org/wiki/Arts_and_Crafts_movement#Criti...</a>
> Any content published by our Lyria model will be watermarked<p>What is the purpose of marking AI content? I don't really understand. Should we mark content edited in GIMP or Audacity as well?
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.<p>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 but at least we'll have another option.
There are still people that appreciate watching someone play a guitar or violin or piano, or make traditional art, because the character of the artist comes through. For example, with classical pieces, each musician plays the composition a little bit differently, and so their character comes through, their soul. I don't think a machine can really replace that, because there are still people that like to listen to live artists play at a classical concert or in a coffeeshop, when that same music has been able to be composed electronically for a couple of decades now.<p>The problem though is that fewer and fewer people are interested in this sort of thing and more interested in generic electronic garbage background music, as others have said. Classical concerts are getting less attendance over time, there are less acoustic or live performances, etc. People altogether seem to no longer buy traditional art because they can buy prints that look good on the wall. The point of buying traditional art isn't just it looking good, because like musicians, the character of the artist and their personality comes through in the paint strokes, color choices, design, composition, etc. Which is why people close to the artist often appreciate the work even more than a stranger would (ie: your 4-old child's school drawing), because the way it's created reminds you of that person and their soul/spirit.