SUNO is killer!

[BONES]

Righto. You're a Social Darwinist. Not a good thing to be, when you're old and fat.

Hey, I've lost 34kg, thanks. I'm now within the average weight range for my height/age. But I've had my time, I have no expectation of living forever and no real desire to be properly old. Dropping dead on stage at our next gig has more appeal to me.

Sony has now a tool that recognizes what tracks the AI music tools stole from, good! Easier to sue AI music creators now.

This woman is just a bottom-feeding ambulance chaser who couldn't lie straight in bed. Just because Sony says they have a tool doesn't mean anything will stand up in court and she f**king knows it. She's just trying to get more greedy, lazy c**ts to sign up to her law suit so she can make more money.

But if you think it is going to stop at AI music if it works, you're kidding yourself. You can expect every slightly similar riff or melody you use to be scruitinised for any unconscious thing you might have accidentally borrowed without realising it. It will f**k everyone over if it can stand up in court, not just AI.

That said, I don't think it will work on any but the most blatant rip-offs anyway. Looking at the 30 or so things my bandmate has sent me in the last few months, I can't identify anything that might have been taken from any song I know. AI may take inspiration form other music, as we all do, but what it creates is no more of a rip-off of its training data than any song you or I write ourselves, unless that's what you've told it to do. I think it's that latter category, where people are deliberately trying to pass off their AI generated songs as coming from a known artist, where tools like this might be effective. i.e. They will work to quantify something that we can all hear for ourselves anyway.

[Zeisner]

Sony has now a tool that recognizes what tracks the AI music tools stole from, good! Easier to sue AI music creators now.

Not before the required legislative acts all around the world. Which is unlikely to happen anytime soon, thanks to the current political climate.

[ksandvik]

You don't need legislative acts for suing private companies.

[Zeisner]

You don't need legislative acts for suing private companies.

For theft you do. Remember stealing samples? Same thing. There must be a legislative act as a foundation.

[ksandvik]

Easy then to use the legistlation of samples and add on top of it copyright protection for anything creative done here in USA - no need to even register it unless you want more solid evidence. Or, best to stay away from Sumo, you never know what happens with material you published via Sumo out there...

[ghettosynth]

Easy then to use the legistlation of samples and add on top of it copyright protection for anything creative done here in USA - no need to even register it unless you want more solid evidence. Or, best to stay away from Sumo, you never know what happens with material you published via Sumo out there...

That isn't going to happen. There has been some pushback, yes, in the courts, of that bright line rule. It was lazy and overly restrictive. You don't even want this even though you think that you do. That rule was the equivalent of being able to copyright short sentences. Imagine getting sued because your KVR posts aren't sufficiently original.

VMG Salsoul, LLC v. Madonna Louise Ciccone, 824 F.3d 871 (9th Cir. 2016):

VMG Salsoul v. Madonna addressed whether a very small, unlicensed digital sample of a sound recording automatically constitutes copyright infringement. The dispute arose from Madonna’s 1990 song “Vogue.” VMG Salsoul, which owned the copyright to the 1977 disco track “Love Break,” alleged that “Vogue” incorporated a 0.23-second horn hit from “Love Break,” sampled and altered multiple times.

The key legal issue concerned whether any unauthorized sampling of a sound recording, no matter how minimal, is per se infringement. The Sixth Circuit had previously adopted a strict bright-line rule in Bridgeport Music, Inc. v. Dimension Films (2005), holding that for sound recordings, “get a license or do not sample.” Under Bridgeport, even trivial copying could trigger liability without any de minimis exception.

The Ninth Circuit declined to follow Bridgeport. It held that the longstanding de minimis doctrine applies to sound recordings just as it does to other copyrighted works. Under this doctrine, copying is not actionable if the portion taken is so small or insignificant that an ordinary observer would not recognize the appropriation. After reviewing the audio evidence, including expert testimony, the court concluded that the sampled horn hit was not recognizable to the average listener and therefore was de minimis.

To me, the most interesting thing about this entire debacle is how much flailing about those who never had any genuine stake in the first place are doing. It's the same across different use cases where scarcity is being challenged.



This is the 10 hour version of the youtube version of this. It has nearly 10k views over the year. The original on Suno has 3.2 million plays, 3k comments, 44k likes. In the comments people are talking about how it's the definitive Suno song. Watch as people form communities around this and exclude the mediocrity that is the overwhelming majority of hobbyist uploads to Spotify.

Note that the music is completely incidental to the expression. The style in Suno is "Cat." This is the quintessential example of how so many here seem to have lost the plot. Nobody gives a shit about the genre, the quality of the synth parts, artifacts, how polished it is, or it isn't. The song is relatable to other people no matter what you think of it. It's a social moment about cats, the internet's first love. That's how people are using this and most KVR members are not only not a part of the conversation, they're not even invited to the table.

[wonshu]

It's a social moment about cats, the internet's first love. That's how people are using this and most KVR members are not only not a part of the conversation, they're not even invited to the table.

That's probably true, at least it's logical.

That "table" was never my table in the first place - so for me personally nothing has changed so far: I still don't care about that "table". We still have to find those people that are equally not interested in that table.

Yes, it's probably going to be much more live (gigs and interaction) and I think that's very likely a good thing for everybody.

[wagtunes]

FWIW, this is my take on Suno or whatever.

First off, if people want to use it, power to them. I don't condemn them for it.
As for the sound, all I know is the current #1 country song was completely made with AI. You may not like it but it obviously appeals to a lot of people.

I don't know what the future of AI is nor do I care, outside of my one little site that I use to do my AI vocals. Everything else I do myself.

Anyway, that's all I got.

[zerocrossing]

You don't need legislative acts for suing private companies.

But you do have to show damages. Can labels show that their music is listened to and streamed less? Can they argue that people aren't just going to a better alternative? Tricky.

[zerocrossing]

I don't know what the future of AI is nor do I care, outside of my one little site that I use to do my AI vocals. Everything else I do myself.

That's not quite what we're talking about. That's basically vocal synthesis, where the engine uses machine learning to create a model that mimics a voice. SWAM instruments also use this to some degree. There's really not much difference. What we're talking about is being able to generate a full song by typing a few sentences. Maybe effective, but I can't imagine a more boring endeavor, but I guess it's good for people who are not creative.

[wagtunes]

I don't know what the future of AI is nor do I care, outside of my one little site that I use to do my AI vocals. Everything else I do myself.

That's not quite what we're talking about. That's basically vocal synthesis, where the engine uses machine learning to create a model that mimics a voice. SWAM instruments also use this to some degree. There's really not much difference. What we're talking about is being able to generate a full song by typing a few sentences. Maybe effective, but I can't imagine a more boring endeavor, but I guess it's good for people who are not creative.

Well, I never had any interest in doing that. However, I have to admit that if I submitted it one of my finished songs (the lyric part) and told it what I wanted, I'm curious as to what it would come up with even if I'd never use it. But since I'd have to pay to use any of these tools worth anything at all, that's not going to happen.

[audiojunkie]

I just read through the whole thread..........whew! What I find interesting potential that no one has mentioned is multisampling. I love samples, and it seems to me that these tools would be ideal for creating multisamples for samplers. Want a saxophone? Tell AI to output samples across the map--one for each key. Doesn't sound right? Upload a sample similar to what you want to hear, and tell it to make the multisample set more like the sample you uploaded. Want articulations? Done. Want loop points? Done. You could even tell AI to output the instrument in the sample format of your choice. This is the kind of thing that I think AI could be really good for. Instrument sounds are usually pretty good in isolation at this technological point. Imagine being able to create your own instrument libraries. This is stuff that could be done already if we wanted to.

There are other useful things that it can do right now that doesn't take away from the creativity, and leaves it in the artists hands:

Imagine taking your vocal track and inputting it into an AI and having it correct the flaws in timing and intonation and polish. This can already be done.

Imagine humming a tune, and then describing the rhythm desired for guitar, including up strokes, downstrokes, mutes, bends, etc. This can already be done.

AI is a tool. Nothing more. To an artist, it can be a great help already at the technological point it is at today. Or, it can be a slop machine. But make no mistake. It is a tool.

[Kenmac]

I just read through the whole thread..........whew! What I find interesting potential that no one has mentioned is multisampling. I love samples, and it seems to me that these tools would be ideal for creating multisamples for samplers. Want a saxophone? Tell AI to output samples across the map--one for each key. Doesn't sound right? Upload a sample similar to what you want to hear, and tell it to make the multisample set more like the sample you uploaded. Want articulations? Done. Want loop points? Done. You could even tell AI to output the instrument in the sample format of your choice. This is the kind of thing that I think AI could be really good for. Instrument sounds are usually pretty good in isolation at this technological point. Imagine being able to create your own instrument libraries. This is stuff that could be done already if we wanted to.

There are other useful things that it can do right now that doesn't take away from the creativity, and leaves it in the artists hands:

Imagine taking your vocal track and inputting it into an AI and having it correct the flaws in timing and intonation and polish. This can already be done.

Imagine humming a tune, and then describing the rhythm desired for guitar, including up strokes, downstrokes, mutes, bends, etc. This can already be done.

AI is a tool. Nothing more. To an artist, it can be a great help already at the technological point it is at today. Or, it can be a slop machine. But make no mistake. It is a tool.

Regarding AI sampling, I don't know if you saw this, but EastWest has partnered with ACE Studio which is an AI DAW.

https://www.kvraudio.com/focus/ace-stud ... tion-65954

[zerocrossing]

I just read through the whole thread..........whew! What I find interesting potential that no one has mentioned is multisampling. I love samples, and it seems to me that these tools would be ideal for creating multisamples for samplers. Want a saxophone? Tell AI to output samples across the map--one for each key. Doesn't sound right? Upload a sample similar to what you want to hear, and tell it to make the multisample set more like the sample you uploaded. Want articulations? Done. Want loop points? Done. You could even tell AI to output the instrument in the sample format of your choice. This is the kind of thing that I think AI could be really good for. Instrument sounds are usually pretty good in isolation at this technological point. Imagine being able to create your own instrument libraries. This is stuff that could be done already if we wanted to.

There are other useful things that it can do right now that doesn't take away from the creativity, and leaves it in the artists hands:

Imagine taking your vocal track and inputting it into an AI and having it correct the flaws in timing and intonation and polish. This can already be done.

Imagine humming a tune, and then describing the rhythm desired for guitar, including up strokes, downstrokes, mutes, bends, etc. This can already be done.

AI is a tool. Nothing more. To an artist, it can be a great help already at the technological point it is at today. Or, it can be a slop machine. But make no mistake. It is a tool.

You're unaware of what AI can actually do. Multisamples are soon going to be completely obsolete. Kontakt... will probably go away. That's probably why NI is in bankruptcy proceedings right now.

With something like ACE Studio, all you have to do is play your MIDI part, or hum it, and tell it to get played back as one of many synthetic instrument models that sound as convincing as your multisampled version, and didn't take you days of putting in all the articulation cues. It doesn't have every instrument, yet, but I don't see myself ever using a sample based instrument to get orchestral sounds ever again. What a tedious PITA. RIP.

[BONES]

First off, Kontakt isn't going anywhere. It's installed on my computer and that's how it will stay for as long as I can keep installing it on new computers. It's a mature product, no-one will care if it never gets beyond the current version, as long as it keeps working. I've got a standalone installer for it, so I'm schweet.

ACE Studio uses East-West multisamples, so you're using samples anyway, you're just paying a shit-tonne of money for them. And ACE Studio seems to have a very narrow focus when it comes to styles, I can't see it being much good for all but the most popular genres. It also looks like you have to put your articulation cues in, too. Have you got a license or are you just speculating?

Maybe effective, but I can't imagine a more boring endeavor, but I guess it's good for people who are not creative.

Give it a go, then, and see how f**king hard it is to get something worthwhile. See how creative you need to be with your prompts to get the result you're after.

Most of you would know by now how much I f**king hate sitting in front of a DAW for countless hours finishing a song but I do it because the results are worth the effort. It's the same with AI - it's a grind but, just like doing it any other way, the results are worth the effort, so you make the effort. Personally, I find the process with AI to be far less onerous than doing it the old way. Maybe that's because I'm not the one doing the prompts but my bandmate loves that stuff, so it works out well for both of us.