17 ms·

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

209 points by kmisiunas 3y ago

kevmo314 3y ago
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. Can you imagine if the GraphCast article were like that? "Revolutionizing the future of weather". I'm sure this music thing is cool but I have no idea what was built. [1]: https://deepmind.google/discover/blog/graphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting/ https://deepmind.google/discover/blog/graphcast-ai-model-for...
- Dalewyn 3y ago
  It's because the audiences are different. Lyria is for operators, Graphcast is for engineers. The former care about what something does, the latter care about how something does.
- armcat 3y ago
  It's definitely a stark contrast, but it's also two vastly different domains. One is traditionally the domain of deep science and simulations, and the other is the domain of artists and creatives. Their announcement is also pretty clear - it's a new generative model (Lyria), which seems to do some impressive vocals generation (at least in the sampled videos); combined with a creators' platform/toolbox.
- jacquesm 3y ago
  For now weather models are - fortunately - not generative.
lynx23 3y ago
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.
- huijzer 3y ago
  Earning money by making music is a power law, that is, only a few people will make a lot of money while many will earn almost nothing. This is partially due to the required skills. If a system could lower the barrier to entry, then you could also make an argument that a music AI makes it a more level playing field.
  - vouaobrasil 3y ago
    That is true, but does it make sense to lower the barrier of entry if in return, the world becomes a fiefdom where Google is part of a tech feudal-lord conglomerate and you have to be subservient to their wishes? Besides, the reason why the power law distribution here is exceptionally skewed is precisely BECAUSE of technology like the internet. If you read about the history of music, in the past before technology, all musicians made less but at least there were places for local musicians to play in a strong community such as in local pubs and such. TECHNOLOGY created the problem of the extreme power law and now it's disguising itself as a solution. But like the devil, a company like Google does not offer anyhting for free.
  - lynx23 3y ago
    Looking at EDM, the sheer amount of bedroom musicians capable of producing that poped up since the 90s is part of the reason why it has become so hard to make any money from your art. There is simply too much music out there to actually be able to find the gems you like. Lowering the barrier for entry will only worsen this effect.
    - helboi4 3y ago
      Very real and pertinent example. Being big in EDM now is 100% about being a constant networker. You have to go to every party, know every person and have a massively professional social media presence and a conisistent flashy image to have any success. This will be accelerated in AI, actually turning being a musician into a ridiculously demanding amalgamation of many many different jobs... or just a game of having enough money to know people already and have employ a PR army.
  - 3y ago
gardenhedge 3y ago
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 but at least we'll have another option.
- iamsaitam 3y ago
  This is sarcasm right?
  - Mashimo 3y ago
    You don't think this will lower the barrier of entry for music creation?
    - vouaobrasil 3y ago
      NO, it will lower the barrier for music generation, which is different from CREATION, which implies a creative act.
      - gardenhedge 3y ago
        I think you're looking at this wrong. If I have an idea for a song I my head, I could record myself humming or sining it. While it might represent a thought, feeling or experience it's not going to be very pleasant. With AI, I will be able to transform it into something enjoyable. That is creation and not just generation.
        Juliate 3y ago
        Between someone that generated music from ideas, and someone that spent time and thinking and practicing to refine and these ideas, each will not relate the same to the end result. If only because the latter will know which happy/sad accidents made it to the end, and which did not, based on a specific judgement, or moment. I've already done both. Generators are only good to produce specific ideas to integrate sometimes, at most. It's exactly the same difference between writing your own novel, and getting it written by a text generator. You don't relate to the story, and to the characters, the same way. You may not even _explain_ them the same way.
light_hue_1 3y ago
Yet another tool from Google that's closed. No model, no code, no real details. You can't even run the model today. Google continues to live in the darkest timeline when it comes to ai and ml.
- vouaobrasil 3y ago
  I certainly hope this will fail, but because it's part of YouTube and thus part of their main advertising goal, it probably won't.
- antupis 3y ago
  Google is living the innovator dilemma on steroids at the moment because Google Search + ads are so profitable everything else is a second thought and Google does not even bother to release stuff that they are developing.
- helboi4 3y ago
  Yet another reason why we should be skeptical about AI and stop these companies from accelerating their monopoly of markets using it. The only way that AI can be good is if we all have equal opportunity to participate in the revolution. Otherwise, we will come to reside in an increasingly technofuedalist state where all smaller businesses are required to pay a tax to exist to the large companies who have their super secret AI models on massive servers that nobody with less capital can compete with, and the average worker will become totally obsolete and destitute.
- sevagh 3y ago
  Reminds me of magenta/ddsp - vaporware.
- butlike 3y ago
  They're just making headlines for shareholders. It's fine.
atleastoptimal 3y ago
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"
- threeseed 3y ago
  Already been seeing this play out in photography. Huge trend back to 90s CCD digital cameras, film, polaroid etc.
- sirwitti 3y ago
  Quality never was the only or even the most important aspect of popularity in music and probably art in general. So I'm not sure your projection is valid.
  - vouaobrasil 3y ago
    His projection is valid because it's not a qualitative problem, it's a quantitative one. Music may already have been turned into a commodity selected by traits other than quality, but tools like this bring the effect up to a whole new level, a whole new order of magnitude.
    - kkzz99 3y ago
      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.
      - vouaobrasil 3y ago
        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 listen to real music created by real people (preferably live) and you will know how hideous these results really are. Everything, even EFFICIENCY, has its limits to how much we should have.
zb3 3y ago
Did they release it? If not, then it's because it sucks, not because they care about "safety".
- vouaobrasil 3y ago
  Google doesn't care about safety. It's exceptionally naive to think that a company that pulls in this kind of money to ever think about safety. They only pay lip service to it to keep their legal department happy. They are leeches.
_def 3y ago
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?
vouaobrasil 3y ago
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 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. > 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. 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.) 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. 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: (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. (2) It changes the minds of human beings to think more like machines. (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. 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. Shame on Google for this and I truly despise them.
- vjerancrnjak 3y ago
  But isn't music more than just the music track? People like to listen to the same music, go to concerts, share it with others, share the experience of audience/artist performance. Moments where music is truly important in one's life will never be replaced by meaningless bit fiddling. I cannot imagine people caring about books, music, art created by bit fiddling. They will always care about things created by other people. We do not watch chess engines playing together, even though the level of play is "godlike". AlphaZero didn't make human chess less popular. Chess is extremely popular now and it is mostly due to personalities participating in it and the ease of access (streaming videos...). I find the demo where a random song from an artist you like is created very weird. Who would want to do that? And why would anyone connect to that? Will these people save these random creations and relisten them? Will they remember these moments by remembering these random silly unique tracks? I cannot imagine that at all.
Almondsetat 3y ago
Will be great for people who have studied composition, for everyone else it will just increase exponentially the amount of garbage circulating around
- codedokode 3y ago
  There is so much garbage already on video sharing apps, I don't think anything will change.
- criddell 3y ago
  Or maybe something entirely unexpected will happen. You never know what can happen when you give people creative tools. Maybe music creation via tools like this will be like using a camera to create images. We don’t think of photographs as shitty paintings, right? It’s a different kind of thing.
  - Almondsetat 3y ago
    > Maybe music creation via tools like this will be like using a camera to create images. Yes, in the sense that most people don’t know how to take photos and they look like crap even with heavy AI retouching
    - criddell 3y ago
      My photos might look like crap but I’m happy to have them. Maybe something similar will happen with AI-powered music creation?
- Musikstueck 3y ago
  It might just be a way for me to create my own art. It would also already work very well for creating background music without bothering artists to do so. Concentration background music etc.
bambax 3y ago
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-a765-0a64650f36f9 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. [0] https://www.youtube.com/watch?v=4-43lLKaqBQ https://www.youtube.com/watch?v=4-43lLKaqBQ
- halflings 3y ago
  Growing pains. Remember how this all started (Google Magenta, circa 2019): https://www.youtube.com/watch?v=ZRnbbtqxBEc https://www.youtube.com/watch?v=ZRnbbtqxBEc
  - bambax 3y ago
    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?
    - Kye 3y ago
      That sounds almost like Scaler: https://www.scalerplugin.com/ https://www.scalerplugin.com/ I mostly use it as a database of chord progressions, but it can do quite a bit more.
    - voidhorse 3y ago
      Its because the ultimate goal with these gen AI tools is to replace not support professionals. Your idea makes sense, is eminently doable, and would be awesome. The problem is, it still requires a human with expertise. Gen AI companies want to remove the human with expertise from the equation precisely so they can "empower" (take money from) the much wider population of humans without expertise. These humans, since they lack expertise will actually need gen AI to produce anything at all. A professional could use it but could ultimately make do without it--they would not have a strong dependency on gen AI. It's essentially the same idea as addicts being lucrative prospects for drug dealers--once you establish dependency, you've got them. Let's be frank, gen AI projects are ultimately about companies wanting make money by putting small-scale professionals out of work for good. In other words, they want to capture a significant portion of all that capital that currently flows across small companies and projects (even if small scale operations continue a large percentage of their production will require funneling money to Gen AI subscriptions instead of skilled human workers)
barrkel 3y ago
This would be like 50x more cool if that example UI you scroll down to was real and live and embedded on the page.
- ilaksh 3y ago
  Yeah. How do we use this? I assume since it's Google it's yet another announcement of a technology that they invented but only a few special people will be able to play with.
codedokode 3y ago
> Any content published by our Lyria model will be watermarked What is the purpose of marking AI content? I don't really understand. Should we mark content edited in GIMP or Audacity as well?
- helboi4 3y ago
  Because it undermines all human achievement to allow one company with their close source AI model to devalue all other products in the market at a speed previously unprecedented to man and honestly incomprehensible (could take literally less than hours), without us being able to monitor it.
  - butlike 3y ago
    What are you scared of?
    - helboi4 3y ago
      Humans becoming basically obsolete and our whole system of society collapsing allowing only the owners of closed source AI systems on incredibly expensive servers to make money, throwing the rest of us into destitution.
stefanos82 3y ago
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.
- dragongling 3y ago
  Nodoby forbids you to create art for the sake of art however you want, be it by recording instruments, making stuff in FL Studio, mixing meme samples or using AI. If you want to be competitive in music production industry you have to be somewhat economically effective as in any industry and AI will raise economical effectiveness to the whole new level.
- mnd999 3y ago
  Totally agree. Fuck off google.
  - hashtag-til 3y ago
    This is becoming Reddit. Please continue.
- jillesvangurp 3y ago
  You are not being deprived of any rights. You just don't have them exclusively. Make music, or don't. Up to you. Either way, there will be machines also making music. So, what?
  - chpatrick 3y ago
    How are people going to become professional musicians if it stops being a career?
    - deleted 3y ago
iamnotsure 3y ago
"Propaganda is making puppets of us. We are moved by hidden strings which the propagandist manipulates."
codedokode 3y ago
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. 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?
siva7 3y ago
I can't wait until youtube content creators start using ai-generated music in their shorts and reels. What a wonderful world.
jamesdhutton 3y ago
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!
- rperez333 3y ago
  This already happens with chess - despite the machine being superhuman, people have never been more attracted to the human game. Perhaps what will happen is that we will have more time to play/watch more?
- wildrhythms 3y ago
  Unfortunately for many people music is nothing more than 'something to play in the background'.
  - datpiff 3y ago
    I'm sure this reads as very cynical, but a lot of the replies here seem to be written by people who do not like music, or don't place much value in it.
    - Supernaut 3y ago
      I think you're correct. I also think that they're envisioning this tool's potential in terms of the kind of derivative, trite rubbish that makes up 95% of what is played on pop radio stations nowadays. I don't doubt that's achievable. What I do very much doubt is that software will ever capable of auto-generating something with the beauty, passion and originality of a Beethoven symphony, as performed by the Berlin Philharmonic under the direction of von Karajan. I don't believe that an algorithm will ever be able to compose and render a new song in the style of Nick Drake, that has even 1% of his soulfulness. Machines can be programmed to imitate patterns of musical notes, or even the timbre of a voice. That's nothing. Lived experience and human emotion are prerequisites for the creation of great art.
seydor 3y ago
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. 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.
- coldtea 3y ago
  >Composer? no. Performer? no. Not even audience, since few care for music with any real passion anymore among the masses (not even teens). It became a niche, or something you put in the background. As for "popular" concerts and festivals, they are more like country fairs...
  - cmoski 3y ago
    There are still a lot of people that care about good music. Most of the shows I go to are sold out and I certainly don't listen to that top 40 trash. I don't see "the masses" there, nor would I want to.
  - phatskat 3y ago
    Maybe I’m in a weird bubble, but I know plenty of people who love music with a passion. We travel for concerts, we go to festivals with lots of other passionate music enjoyers, heck we have our own yearly get together that’s about playing music with friends and family. The people I know who are into music are into music, and that’s most of my friends because those are the people I get along with - we make, play, and listen with passion, and there are a lot of us.
  - butlike 3y ago
    I think you might be applying your perspective a little globally. For lots, music is NOT something that's necessarily put in the background.
gverrilla 3y ago
Google AI can't even produce proper subtitles yet for youtube.
ppsreejith 3y ago
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.
- mavamaarten 3y ago
  Right? The only thing I fear for is the amount of generic AI generated trash that will be spawned among good music. But I'm super excited about this honestly, I only play guitar and have a musical mind but am absolutely horrible at production tools. Something like this could totally let me create a real production of what I imagine/hum/sing, which would otherwise be impossible.
- Juliate 3y ago
  It's not that the choice is taken away. It's that it is going to be even more buried under an avalanche of industry-normalised, generated crap (crap because generated and industry-normalised). And second, you express yourself, you said it. If a generator produces something, it's not _your_ expression, per definition. You may still have "constrained" it among myriad of possibilities through your prompt, it is still the production of the generator, not yours. Example: when I play something on a Mellotron, I cannot claim that I played the flutes or violins recorded there. Those are definitely not mine. What I do play with it, what I sing above it, yes, maybe because I produce/play those, with what I understood of my previous musical education/experience. That's, fundamentally different.
  - ppsreejith 3y ago
    Regarding the first point of being buried under an avalanche of crap, we can solve this problem through many ways. We can still filter by human music. Or we can use better recommendation systems (automated or p2p/word-of-mouth). Regarding whether it is "my" expression or not, I consider expression as a way of making my feelings or thoughts known. This grants me an additional way of making my feelings or thoughts known. Perhaps of lower fidelity than if I learn to play an instrument but maybe that will improve with time. Either way, since I now have an additional choice of expressing my thoughts and _I_ decide if it's "good enough" to publish, I consider it as allowing me to express more.
Musikstueck 3y ago
What we do as humans is to push against a wall of mist of a unlimited search space. Doesn't matter if it is music, drawing or scientific discoveries. What an Artist does is pushing the boundaries. Having tools like ths allows much more people to push and not 'just' the artists who mastered certain skills. 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. 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. But hopefully we will have more time to learn all of those skills for the sake of having fun learning them. 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? don't be that type of person
- InCityDreams 3y ago
  >Doesn't matter if it is music, drawing or scientific discoveries. #oxford comma blues.
  - netrus 3y ago
    As drawing discoveries are hardly a thing, I see no ambiguity here.
  - Musikstueck 3y ago
    Im german, had the second worst grade in english when i graduated. Feel free to be more precise than a hashtag :P
KaiserPro 3y ago
To quote from 1984: > 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. Part of me worries that I am just gatekeeping. 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? 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. 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? 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] I also frankly worry about how my kids are going to make money. [1]https://en.wikipedia.org/wiki/Arts_and_Crafts_movement#Critique_of_industry https://en.wikipedia.org/wiki/Arts_and_Crafts_movement#Criti...
- phatskat 3y ago
  > I also frankly worry about how my kids are going to make money. Then seriously start thinking about which politicians are going to support a future where fewer and fewer jobs are being done by humans. UBI, healthcare for everyone, access to housing etc are all things that can and should be done already. The more that work is automated away, the more crucial these become. We should be striving for a “Jetsons future” where complaining about a long work day means being in the office for two hours.
  - ethbr1 3y ago
ristos 3y ago
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. 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.
- ristos 3y ago
  Not sure why I got downvoted. There's a false dichotomy here... I agree that there are really cool things with this technology, opening up a lot of possibilities for music creation for those that aren't well versed in music composition to get their ideas out. I think that's great. Just like how electronic music allowed us to have digital recordings to play the same song cheaply over and over again, and also allowed people to create great tracks with only knowledge of music composition and no need to learn an instrument. That still doesn't change some drawbacks, which is what you lose when you do that. It's mostly an economics thing. There's value to having artisanal products over industrial ones because of the craftsmanship involved. The artist or craftsman's personal touch ends up as part of the piece. That used to be common but now it's just a high end product for rich people to afford. There's a downside to these kinds of technologies, even if they bring benefits.
  - dave333 3y ago
theGnuMe 3y ago
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. Imagine the talent AI assisted music programs will unlock. I think it will be a new instrument people learn to play.
darkerside 3y ago
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).
kmfrk 3y ago
> 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. 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.
- richrichardsson 3y ago
  This whole section made me laugh. Maybe it was AI generated, because "converting the audio wave into a two-dimensional visualization that shows how the spectrum of frequencies in a sound evolves over time" is far from a "novel method (is) unlike anything that exists today, especially in the context of audio". Perhaps they're talking about the actual watermarking method, but it's a terrible pair of sentences if so.
pcdoodle 3y ago
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.
glimshe 3y ago
I'd like a version of this that can do MIDI, like the next generation Band-in-a-box.
simonebrunozzi 3y ago
How can I use Lyria?
marviel 3y ago
Art is going to become less about creating beauty, and more about transmitting information.
- anonbuddy 3y ago
  could you elaborate more on that statement and how you envision the future of art? sounds interesting
  - marviel 3y ago
    It's an intuition biased on my own preferences, and a bit of a "Misnomer" because ideas can also be beautiful, but: IF we assume that: 1. "Beautiful" (nice "looking" / "sounding" / "moving" things) things will be easy to create. 2. "Meaningful" things are, and will continue to be, harder to create.[A] Then it seems that the only thing that could differentiate you is the quality of your ideas, and the meaning you are transmitting. Put another way -- if everyone can create "Beautiful" things by default (in the world of static art, we live here now!) then why choose one piece of art over another? This is my intuition about it, at least. ----- [A]: I think this is just a fundamental property of "meaningful" things. They require more computation, which means more scarcity.
dwringer 3y ago
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 part of a song (which is arguably considerable as fair use of even copyrighted material if only a single bar is sampled), and the statement: > 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*. So... we're watching you and we'll know if you so much as sample one bar to create something of your own. But our preferred creators are having so much fun using AI exclusively to make the entire composition, directly ripping off the personal style of these artists who said it's OK! 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... * 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.
iterateoften 3y ago
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. 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 connection with the audience. His concert was incredible to experience a room of 1000 people all having their seats shaken 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. 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. 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. 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. 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. 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). 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.
belter 3y ago
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...
zaptrem 3y ago
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.
SubiculumCode 3y ago
[flagged]
gonzaloalvarez 3y ago
[flagged]
jmathai 3y ago
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. To each their own (I mean that) but I do wonder sometimes…At some point, what’s the point?
- jchw 3y ago
  Some of this generative AI stuff is going to raise some interesting questions. Right now a lot of it is rough around the edges and distinctly inhuman. However, I do believe that even if we're not entirely on the right track, there's not much reason to believe a machine can't do human-like things as well as humans with enough computational power. Here's my guess: I think that this stuff will continue to iterate until it's to the point where the amount of control people have over it as well as the quality of handling edge cases is sufficient to allow people with fairly unsophisticated tools to create things that look fairly high quality. And once this happens, people armed with amazing tools but no better taste than they had before, will flood the Internet with a bunch of tasteless garbage at rates never before seen. It hasn't happened yet, but I think this is partly an accessibility issue (not everyone has datacenter GPU access, there's still limitations in open models, and many hosted offerings are pretty limited compared to what local models can do right now in terms of tooling) but also it will make a big difference if we hit the point where it's really genuinely just not possible to reliably distinguish AI generated images and sound. Then, even if people wanted to ignore AI generated media, it would probably quickly become pretty difficult, and many people will probably not be honest if it limits their reach and potential.
  - DrSiemer 3y ago
    It will just raise the bar of entry. Tolerance for generic garbage will decrease. Elevator muzak producers are in trouble. Real talent will still float to the top.
    - diydsp 3y ago
codekansas 3y ago
It's really interesting for me to see AI voice cloning taking off. A few years ago I worked on a paper called HuBERT (https://arxiv.org/abs/2106.07447 https://arxiv.org/abs/2106.07447) 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.
nextworddev 3y ago
I'm guessing Charlie Puth is getting some revenue share.
DevX101 3y ago
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.
- DrSiemer 3y ago
  The requirement to either pay a lot for or learn how to do mastering always seemed unnecessary to me. Stuff like mixing, equalizing, sidechaining and eternal sample hunts all take a lot of time as well... It's why making music is one of the hobbies I had to let go. Can't wait to get my hands on an AI powered DAW that skips all the tedious stuff! Am I still the one making the music? Well, I'd argue the 300+ hours I'll probably sink into finetuning all the other stuff will say yes.
  - djkoolaide 3y ago
    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 :)
    - bigbluedots 3y ago
      Where can one find multitracks on the internet?
      - Pyrodogg 3y ago
        Cambridge Music Technology has a great list of multi tracks for learning the art of mixing. The submissions cover many genres. https://www.cambridge-mt.com/ms/mtk/ https://www.cambridge-mt.com/ms/mtk/
darklycan51 3y ago
[flagged]
darklycan51 3y ago
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? 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". 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? One thing I know is that whatever Google says is a lie.
Davidzheng 3y ago
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.
interestica 3y ago
> 9 artists ... who have chosen to collaborate in this experiment and work with us to shape the future of AI in music I really hope it's not google that drives the future of AI in music.
- squigz 3y ago
  They will be if people let them.
orliesaurus 3y ago
Why did they pick T-Pain for their demo, out of all the good musicians in the world?
- djmips 3y ago
  He embraced applying tech to music early and he's not the terrible musician you might be thinking.
vsnf 3y ago
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.
TylerJaacks 3y ago
This is cool, I'm sure people will use this responsibly and not take advantage of this.
hollowturtle 3y ago
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.
dang 3y ago
Related: https://blog.youtube/inside-youtube/ai-and-music-experiment/ https://blog.youtube/inside-youtube/ai-and-music-experiment/ (via https://news.ycombinator.com/item?id=38298670 https://news.ycombinator.com/item?id=38298670, but we merged the comments hither)
barbariangrunge 3y ago
Because musicians haven't been hurt badly enough already by big tech
iainctduncan 3y ago
To 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.
- kniffy 3y ago
  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.
  - iainctduncan 3y ago
    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!

209 points by kmisiunas 3y ago

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. Can you imagine if the GraphCast article were like that? "Revolutionizing the future of weather". I'm sure this music thing is cool but I have no idea what was built. [1]: https://deepmind.google/discover/blog/graphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting/ https://deepmind.google/discover/blog/graphcast-ai-model-for...

Dalewyn 3y ago
It's because the audiences are different. Lyria is for operators, Graphcast is for engineers. The former care about what something does, the latter care about how something does.
armcat 3y ago
It's definitely a stark contrast, but it's also two vastly different domains. One is traditionally the domain of deep science and simulations, and the other is the domain of artists and creatives. Their announcement is also pretty clear - it's a new generative model (Lyria), which seems to do some impressive vocals generation (at least in the sampled videos); combined with a creators' platform/toolbox.
jacquesm 3y ago
For now weather models are - fortunately - not generative.

lynx23 3y ago

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.

huijzer 3y ago
Earning money by making music is a power law, that is, only a few people will make a lot of money while many will earn almost nothing. This is partially due to the required skills. If a system could lower the barrier to entry, then you could also make an argument that a music AI makes it a more level playing field.
- vouaobrasil 3y ago
  That is true, but does it make sense to lower the barrier of entry if in return, the world becomes a fiefdom where Google is part of a tech feudal-lord conglomerate and you have to be subservient to their wishes? Besides, the reason why the power law distribution here is exceptionally skewed is precisely BECAUSE of technology like the internet. If you read about the history of music, in the past before technology, all musicians made less but at least there were places for local musicians to play in a strong community such as in local pubs and such. TECHNOLOGY created the problem of the extreme power law and now it's disguising itself as a solution. But like the devil, a company like Google does not offer anyhting for free.
- lynx23 3y ago
  Looking at EDM, the sheer amount of bedroom musicians capable of producing that poped up since the 90s is part of the reason why it has become so hard to make any money from your art. There is simply too much music out there to actually be able to find the gems you like. Lowering the barrier for entry will only worsen this effect.
  - helboi4 3y ago
    Very real and pertinent example. Being big in EDM now is 100% about being a constant networker. You have to go to every party, know every person and have a massively professional social media presence and a conisistent flashy image to have any success. This will be accelerated in AI, actually turning being a musician into a ridiculously demanding amalgamation of many many different jobs... or just a game of having enough money to know people already and have employ a PR army.
- 3y ago

gardenhedge 3y ago

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 but at least we'll have another option.

iamsaitam 3y ago
This is sarcasm right?
- Mashimo 3y ago
  You don't think this will lower the barrier of entry for music creation?
  - vouaobrasil 3y ago
    NO, it will lower the barrier for music generation, which is different from CREATION, which implies a creative act.
    - gardenhedge 3y ago
      I think you're looking at this wrong. If I have an idea for a song I my head, I could record myself humming or sining it. While it might represent a thought, feeling or experience it's not going to be very pleasant. With AI, I will be able to transform it into something enjoyable. That is creation and not just generation.
      - Juliate 3y ago
        Between someone that generated music from ideas, and someone that spent time and thinking and practicing to refine and these ideas, each will not relate the same to the end result. If only because the latter will know which happy/sad accidents made it to the end, and which did not, based on a specific judgement, or moment. I've already done both. Generators are only good to produce specific ideas to integrate sometimes, at most. It's exactly the same difference between writing your own novel, and getting it written by a text generator. You don't relate to the story, and to the characters, the same way. You may not even _explain_ them the same way.

light_hue_1 3y ago

Yet another tool from Google that's closed. No model, no code, no real details. You can't even run the model today. Google continues to live in the darkest timeline when it comes to ai and ml.

vouaobrasil 3y ago
I certainly hope this will fail, but because it's part of YouTube and thus part of their main advertising goal, it probably won't.
antupis 3y ago
Google is living the innovator dilemma on steroids at the moment because Google Search + ads are so profitable everything else is a second thought and Google does not even bother to release stuff that they are developing.
helboi4 3y ago
Yet another reason why we should be skeptical about AI and stop these companies from accelerating their monopoly of markets using it. The only way that AI can be good is if we all have equal opportunity to participate in the revolution. Otherwise, we will come to reside in an increasingly technofuedalist state where all smaller businesses are required to pay a tax to exist to the large companies who have their super secret AI models on massive servers that nobody with less capital can compete with, and the average worker will become totally obsolete and destitute.
sevagh 3y ago
Reminds me of magenta/ddsp - vaporware.
butlike 3y ago
They're just making headlines for shareholders. It's fine.

atleastoptimal 3y ago

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"

threeseed 3y ago
Already been seeing this play out in photography. Huge trend back to 90s CCD digital cameras, film, polaroid etc.
sirwitti 3y ago
Quality never was the only or even the most important aspect of popularity in music and probably art in general. So I'm not sure your projection is valid.
- vouaobrasil 3y ago
  His projection is valid because it's not a qualitative problem, it's a quantitative one. Music may already have been turned into a commodity selected by traits other than quality, but tools like this bring the effect up to a whole new level, a whole new order of magnitude.
  - kkzz99 3y ago
    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.
    - vouaobrasil 3y ago
      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 listen to real music created by real people (preferably live) and you will know how hideous these results really are. Everything, even EFFICIENCY, has its limits to how much we should have.

zb3 3y ago

Did they release it? If not, then it's because it sucks, not because they care about "safety".

vouaobrasil 3y ago
Google doesn't care about safety. It's exceptionally naive to think that a company that pulls in this kind of money to ever think about safety. They only pay lip service to it to keep their legal department happy. They are leeches.

_def 3y ago

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?

vouaobrasil 3y ago

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 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. > 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. 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.) 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. 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: (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. (2) It changes the minds of human beings to think more like machines. (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. 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. Shame on Google for this and I truly despise them.

vjerancrnjak 3y ago
But isn't music more than just the music track? People like to listen to the same music, go to concerts, share it with others, share the experience of audience/artist performance. Moments where music is truly important in one's life will never be replaced by meaningless bit fiddling. I cannot imagine people caring about books, music, art created by bit fiddling. They will always care about things created by other people. We do not watch chess engines playing together, even though the level of play is "godlike". AlphaZero didn't make human chess less popular. Chess is extremely popular now and it is mostly due to personalities participating in it and the ease of access (streaming videos...). I find the demo where a random song from an artist you like is created very weird. Who would want to do that? And why would anyone connect to that? Will these people save these random creations and relisten them? Will they remember these moments by remembering these random silly unique tracks? I cannot imagine that at all.

Almondsetat 3y ago

Will be great for people who have studied composition, for everyone else it will just increase exponentially the amount of garbage circulating around

codedokode 3y ago
There is so much garbage already on video sharing apps, I don't think anything will change.
criddell 3y ago
Or maybe something entirely unexpected will happen. You never know what can happen when you give people creative tools. Maybe music creation via tools like this will be like using a camera to create images. We don’t think of photographs as shitty paintings, right? It’s a different kind of thing.
- Almondsetat 3y ago
  > Maybe music creation via tools like this will be like using a camera to create images. Yes, in the sense that most people don’t know how to take photos and they look like crap even with heavy AI retouching
  - criddell 3y ago
    My photos might look like crap but I’m happy to have them. Maybe something similar will happen with AI-powered music creation?
Musikstueck 3y ago
It might just be a way for me to create my own art. It would also already work very well for creating background music without bothering artists to do so. Concentration background music etc.

bambax 3y ago

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-a765-0a64650f36f9 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. [0] https://www.youtube.com/watch?v=4-43lLKaqBQ https://www.youtube.com/watch?v=4-43lLKaqBQ

halflings 3y ago
Growing pains. Remember how this all started (Google Magenta, circa 2019): https://www.youtube.com/watch?v=ZRnbbtqxBEc https://www.youtube.com/watch?v=ZRnbbtqxBEc
- bambax 3y ago
  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?
  - Kye 3y ago
    That sounds almost like Scaler: https://www.scalerplugin.com/ https://www.scalerplugin.com/ I mostly use it as a database of chord progressions, but it can do quite a bit more.
  - voidhorse 3y ago
    Its because the ultimate goal with these gen AI tools is to replace not support professionals. Your idea makes sense, is eminently doable, and would be awesome. The problem is, it still requires a human with expertise. Gen AI companies want to remove the human with expertise from the equation precisely so they can "empower" (take money from) the much wider population of humans without expertise. These humans, since they lack expertise will actually need gen AI to produce anything at all. A professional could use it but could ultimately make do without it--they would not have a strong dependency on gen AI. It's essentially the same idea as addicts being lucrative prospects for drug dealers--once you establish dependency, you've got them. Let's be frank, gen AI projects are ultimately about companies wanting make money by putting small-scale professionals out of work for good. In other words, they want to capture a significant portion of all that capital that currently flows across small companies and projects (even if small scale operations continue a large percentage of their production will require funneling money to Gen AI subscriptions instead of skilled human workers)

barrkel 3y ago

This would be like 50x more cool if that example UI you scroll down to was real and live and embedded on the page.

ilaksh 3y ago
Yeah. How do we use this? I assume since it's Google it's yet another announcement of a technology that they invented but only a few special people will be able to play with.

codedokode 3y ago

> Any content published by our Lyria model will be watermarked What is the purpose of marking AI content? I don't really understand. Should we mark content edited in GIMP or Audacity as well?

helboi4 3y ago
Because it undermines all human achievement to allow one company with their close source AI model to devalue all other products in the market at a speed previously unprecedented to man and honestly incomprehensible (could take literally less than hours), without us being able to monitor it.
- butlike 3y ago
  What are you scared of?
  - helboi4 3y ago
    Humans becoming basically obsolete and our whole system of society collapsing allowing only the owners of closed source AI systems on incredibly expensive servers to make money, throwing the rest of us into destitution.

stefanos82 3y ago

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.

dragongling 3y ago
Nodoby forbids you to create art for the sake of art however you want, be it by recording instruments, making stuff in FL Studio, mixing meme samples or using AI. If you want to be competitive in music production industry you have to be somewhat economically effective as in any industry and AI will raise economical effectiveness to the whole new level.
mnd999 3y ago
Totally agree. Fuck off google.
- hashtag-til 3y ago
  This is becoming Reddit. Please continue.
jillesvangurp 3y ago
You are not being deprived of any rights. You just don't have them exclusively. Make music, or don't. Up to you. Either way, there will be machines also making music. So, what?
- chpatrick 3y ago
  How are people going to become professional musicians if it stops being a career?
  - deleted 3y ago

iamnotsure 3y ago

"Propaganda is making puppets of us. We are moved by hidden strings which the propagandist manipulates."

codedokode 3y ago

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. 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?

siva7 3y ago

I can't wait until youtube content creators start using ai-generated music in their shorts and reels. What a wonderful world.

jamesdhutton 3y ago

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!

rperez333 3y ago
This already happens with chess - despite the machine being superhuman, people have never been more attracted to the human game. Perhaps what will happen is that we will have more time to play/watch more?
wildrhythms 3y ago
Unfortunately for many people music is nothing more than 'something to play in the background'.
- datpiff 3y ago
  I'm sure this reads as very cynical, but a lot of the replies here seem to be written by people who do not like music, or don't place much value in it.
  - Supernaut 3y ago
    I think you're correct. I also think that they're envisioning this tool's potential in terms of the kind of derivative, trite rubbish that makes up 95% of what is played on pop radio stations nowadays. I don't doubt that's achievable. What I do very much doubt is that software will ever capable of auto-generating something with the beauty, passion and originality of a Beethoven symphony, as performed by the Berlin Philharmonic under the direction of von Karajan. I don't believe that an algorithm will ever be able to compose and render a new song in the style of Nick Drake, that has even 1% of his soulfulness. Machines can be programmed to imitate patterns of musical notes, or even the timbre of a voice. That's nothing. Lived experience and human emotion are prerequisites for the creation of great art.

seydor 3y ago

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. 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.

coldtea 3y ago
>Composer? no. Performer? no. Not even audience, since few care for music with any real passion anymore among the masses (not even teens). It became a niche, or something you put in the background. As for "popular" concerts and festivals, they are more like country fairs...
- cmoski 3y ago
  There are still a lot of people that care about good music. Most of the shows I go to are sold out and I certainly don't listen to that top 40 trash. I don't see "the masses" there, nor would I want to.
- phatskat 3y ago
  Maybe I’m in a weird bubble, but I know plenty of people who love music with a passion. We travel for concerts, we go to festivals with lots of other passionate music enjoyers, heck we have our own yearly get together that’s about playing music with friends and family. The people I know who are into music are into music, and that’s most of my friends because those are the people I get along with - we make, play, and listen with passion, and there are a lot of us.
- butlike 3y ago
  I think you might be applying your perspective a little globally. For lots, music is NOT something that's necessarily put in the background.

gverrilla 3y ago

Google AI can't even produce proper subtitles yet for youtube.

ppsreejith 3y ago

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.

mavamaarten 3y ago
Right? The only thing I fear for is the amount of generic AI generated trash that will be spawned among good music. But I'm super excited about this honestly, I only play guitar and have a musical mind but am absolutely horrible at production tools. Something like this could totally let me create a real production of what I imagine/hum/sing, which would otherwise be impossible.
Juliate 3y ago
It's not that the choice is taken away. It's that it is going to be even more buried under an avalanche of industry-normalised, generated crap (crap because generated and industry-normalised). And second, you express yourself, you said it. If a generator produces something, it's not _your_ expression, per definition. You may still have "constrained" it among myriad of possibilities through your prompt, it is still the production of the generator, not yours. Example: when I play something on a Mellotron, I cannot claim that I played the flutes or violins recorded there. Those are definitely not mine. What I do play with it, what I sing above it, yes, maybe because I produce/play those, with what I understood of my previous musical education/experience. That's, fundamentally different.
- ppsreejith 3y ago
  Regarding the first point of being buried under an avalanche of crap, we can solve this problem through many ways. We can still filter by human music. Or we can use better recommendation systems (automated or p2p/word-of-mouth). Regarding whether it is "my" expression or not, I consider expression as a way of making my feelings or thoughts known. This grants me an additional way of making my feelings or thoughts known. Perhaps of lower fidelity than if I learn to play an instrument but maybe that will improve with time. Either way, since I now have an additional choice of expressing my thoughts and _I_ decide if it's "good enough" to publish, I consider it as allowing me to express more.

Musikstueck 3y ago

What we do as humans is to push against a wall of mist of a unlimited search space. Doesn't matter if it is music, drawing or scientific discoveries. What an Artist does is pushing the boundaries. Having tools like ths allows much more people to push and not 'just' the artists who mastered certain skills. 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. 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. But hopefully we will have more time to learn all of those skills for the sake of having fun learning them. 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? don't be that type of person

InCityDreams 3y ago
>Doesn't matter if it is music, drawing or scientific discoveries. #oxford comma blues.
- netrus 3y ago
  As drawing discoveries are hardly a thing, I see no ambiguity here.
- Musikstueck 3y ago
  Im german, had the second worst grade in english when i graduated. Feel free to be more precise than a hashtag :P

KaiserPro 3y ago

To quote from 1984: > 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. Part of me worries that I am just gatekeeping. 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? 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. 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? 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] I also frankly worry about how my kids are going to make money. [1]https://en.wikipedia.org/wiki/Arts_and_Crafts_movement#Critique_of_industry https://en.wikipedia.org/wiki/Arts_and_Crafts_movement#Criti...

phatskat 3y ago
> I also frankly worry about how my kids are going to make money. Then seriously start thinking about which politicians are going to support a future where fewer and fewer jobs are being done by humans. UBI, healthcare for everyone, access to housing etc are all things that can and should be done already. The more that work is automated away, the more crucial these become. We should be striving for a “Jetsons future” where complaining about a long work day means being in the office for two hours.
- ethbr1 3y ago

ristos 3y ago

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. 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.

ristos 3y ago
Not sure why I got downvoted. There's a false dichotomy here... I agree that there are really cool things with this technology, opening up a lot of possibilities for music creation for those that aren't well versed in music composition to get their ideas out. I think that's great. Just like how electronic music allowed us to have digital recordings to play the same song cheaply over and over again, and also allowed people to create great tracks with only knowledge of music composition and no need to learn an instrument. That still doesn't change some drawbacks, which is what you lose when you do that. It's mostly an economics thing. There's value to having artisanal products over industrial ones because of the craftsmanship involved. The artist or craftsman's personal touch ends up as part of the piece. That used to be common but now it's just a high end product for rich people to afford. There's a downside to these kinds of technologies, even if they bring benefits.
- dave333 3y ago

theGnuMe 3y ago

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. Imagine the talent AI assisted music programs will unlock. I think it will be a new instrument people learn to play.

darkerside 3y ago

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).

kmfrk 3y ago

> 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. 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.

richrichardsson 3y ago
This whole section made me laugh. Maybe it was AI generated, because "converting the audio wave into a two-dimensional visualization that shows how the spectrum of frequencies in a sound evolves over time" is far from a "novel method (is) unlike anything that exists today, especially in the context of audio". Perhaps they're talking about the actual watermarking method, but it's a terrible pair of sentences if so.

pcdoodle 3y ago

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.

glimshe 3y ago

I'd like a version of this that can do MIDI, like the next generation Band-in-a-box.

simonebrunozzi 3y ago

How can I use Lyria?

marviel 3y ago

Art is going to become less about creating beauty, and more about transmitting information.

anonbuddy 3y ago
could you elaborate more on that statement and how you envision the future of art? sounds interesting
- marviel 3y ago
  It's an intuition biased on my own preferences, and a bit of a "Misnomer" because ideas can also be beautiful, but: IF we assume that: 1. "Beautiful" (nice "looking" / "sounding" / "moving" things) things will be easy to create. 2. "Meaningful" things are, and will continue to be, harder to create.[A] Then it seems that the only thing that could differentiate you is the quality of your ideas, and the meaning you are transmitting. Put another way -- if everyone can create "Beautiful" things by default (in the world of static art, we live here now!) then why choose one piece of art over another? This is my intuition about it, at least. ----- [A]: I think this is just a fundamental property of "meaningful" things. They require more computation, which means more scarcity.

dwringer 3y ago

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 part of a song (which is arguably considerable as fair use of even copyrighted material if only a single bar is sampled), and the statement: > 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*. So... we're watching you and we'll know if you so much as sample one bar to create something of your own. But our preferred creators are having so much fun using AI exclusively to make the entire composition, directly ripping off the personal style of these artists who said it's OK! 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... * 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.

iterateoften 3y ago

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. 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 connection with the audience. His concert was incredible to experience a room of 1000 people all having their seats shaken 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. 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. 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. 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. 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. 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). 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.

belter 3y ago

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...

zaptrem 3y ago

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.

SubiculumCode 3y ago

[flagged]

gonzaloalvarez 3y ago

[flagged]

jmathai 3y ago

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. To each their own (I mean that) but I do wonder sometimes…At some point, what’s the point?

jchw 3y ago
Some of this generative AI stuff is going to raise some interesting questions. Right now a lot of it is rough around the edges and distinctly inhuman. However, I do believe that even if we're not entirely on the right track, there's not much reason to believe a machine can't do human-like things as well as humans with enough computational power. Here's my guess: I think that this stuff will continue to iterate until it's to the point where the amount of control people have over it as well as the quality of handling edge cases is sufficient to allow people with fairly unsophisticated tools to create things that look fairly high quality. And once this happens, people armed with amazing tools but no better taste than they had before, will flood the Internet with a bunch of tasteless garbage at rates never before seen. It hasn't happened yet, but I think this is partly an accessibility issue (not everyone has datacenter GPU access, there's still limitations in open models, and many hosted offerings are pretty limited compared to what local models can do right now in terms of tooling) but also it will make a big difference if we hit the point where it's really genuinely just not possible to reliably distinguish AI generated images and sound. Then, even if people wanted to ignore AI generated media, it would probably quickly become pretty difficult, and many people will probably not be honest if it limits their reach and potential.
- DrSiemer 3y ago
  It will just raise the bar of entry. Tolerance for generic garbage will decrease. Elevator muzak producers are in trouble. Real talent will still float to the top.
  - diydsp 3y ago

codekansas 3y ago

It's really interesting for me to see AI voice cloning taking off. A few years ago I worked on a paper called HuBERT (https://arxiv.org/abs/2106.07447 https://arxiv.org/abs/2106.07447) 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.

nextworddev 3y ago

I'm guessing Charlie Puth is getting some revenue share.

DevX101 3y ago

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.

DrSiemer 3y ago
The requirement to either pay a lot for or learn how to do mastering always seemed unnecessary to me. Stuff like mixing, equalizing, sidechaining and eternal sample hunts all take a lot of time as well... It's why making music is one of the hobbies I had to let go. Can't wait to get my hands on an AI powered DAW that skips all the tedious stuff! Am I still the one making the music? Well, I'd argue the 300+ hours I'll probably sink into finetuning all the other stuff will say yes.
- djkoolaide 3y ago
  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 :)
  - bigbluedots 3y ago
    Where can one find multitracks on the internet?
    - Pyrodogg 3y ago
      Cambridge Music Technology has a great list of multi tracks for learning the art of mixing. The submissions cover many genres. https://www.cambridge-mt.com/ms/mtk/ https://www.cambridge-mt.com/ms/mtk/

darklycan51 3y ago

[flagged]

darklycan51 3y ago

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? 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". 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? One thing I know is that whatever Google says is a lie.

Davidzheng 3y ago

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.

interestica 3y ago

> 9 artists ... who have chosen to collaborate in this experiment and work with us to shape the future of AI in music I really hope it's not google that drives the future of AI in music.

squigz 3y ago
They will be if people let them.

orliesaurus 3y ago

Why did they pick T-Pain for their demo, out of all the good musicians in the world?

djmips 3y ago
He embraced applying tech to music early and he's not the terrible musician you might be thinking.

vsnf 3y ago

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.

TylerJaacks 3y ago

This is cool, I'm sure people will use this responsibly and not take advantage of this.

hollowturtle 3y ago

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.

dang 3y ago

Related: https://blog.youtube/inside-youtube/ai-and-music-experiment/ https://blog.youtube/inside-youtube/ai-and-music-experiment/ (via https://news.ycombinator.com/item?id=38298670 https://news.ycombinator.com/item?id=38298670, but we merged the comments hither)

barbariangrunge 3y ago

Because musicians haven't been hurt badly enough already by big tech

iainctduncan 3y ago

To 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.

kniffy 3y ago
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.
- iainctduncan 3y ago
  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!