I’m not talking about actual music made with Synth V db’s as the vocal because those are fine. What isn’t fine though is this idea that AI Art and AI Music can somehow coexist with Synth V. They’re different things ethically and it makes me feel really weird that DT seemingly allows that stuff to be on the forum. AI Art’s not art. AI Music’s not music. Synth V’s good because it’s ethical on all sides from the voice providers to the way it’s supposed to be used on songs. AI Art and AI Music models take material from millions of sources without permission and puts it together into something that resembles music but lacks any of the heart and soul, especially since nobody consented to having their music or art be taken and used in the dataset. If you can’t do art or music otherwise then either commission someone to do it for you, use royalty free props or learn to do it yourself. There are other ways through that don’t require you to generate things. Basically all I’m asking is for DT to set guidelines for what is and isn’t ethical to put on the forum. Synth V? Absolutely. AI Art? No. AI Music? No. I know I’m probably gonna get some pushback but I wouldn’t be doing this if I wasn’t concerned about how it could look on DT to allow stuff like this on the forum.
For me personally, I think it would be good to have a labeling requirement of how AI was used in the creation of your song, and also require that audio uploads use Synthesizer-V generated vocals. But I don’t think banning the use of AI is necessary. While Synthesizer-V, can be used by professionals to create professional tracks, the majority of people using it will be hobbiests with minimal resources. So, what if you’re great at writing the music, but you suck at writing lyrics? Well, in the well connected world, you’d just connect with a good lyricist. For your average bedroom producer, that’s just not going to happen. Similarly, what if you write amazing lyrics, but you’re dreadful at writing a tune? All you’re doing is denying accessibility to a subsection of the community. And while there may be limited point in critiquing an AI generated song, if the user is looking for a critique, they are as likely to want to improve it based on those critiques, as they are just wanting a click. I’ve seen this issue handled a couple of ways on forums - one banning all use of AI outright, and another saying… lets wait and see if it becomes a problem (it really hasn’t so far).
I’ve used AI in lots of my work, and while I prefer to write my own lyrics and my own music, and do all the production myself, there are times when I want some help. I’ve always been pretty bad at writing lyrics, and I’ve done a number of songs that start out using Chat GPT. Typically I’d use many prompts in order to get the result I’m looking for, and tweak the results to suit my taste in story telling. I don’t simply plug in a title and let it do all its own stuff. I have done that for lyrics, and it quickly became obvious that ChatGPT gets transfixed on something, and struggles to get out of a rut. Or at least it used to. Anyway, it trained me to think more deeply, and explore more vocabulary for word choices for a given subject. It made me able to write, what I think at least, are decent lyrics… and if not decent, at least better than I used to be able to. I think I can write a good melody, and I’m decent at the chord stuff. Production quality is limited by my old ears. In any event, as a hobbiest, you can use AI and still be fully engaged in the process. I think I’ve learned a lot through using gen AI to help me with lyrics, and I’m able to write much better stuff on my own now, than I used to be able to do.
The problem is not even trying to be engaged, but I suspect the people that do that, quickly become bored and move on to another hobby, so the issue should be a minimal one, especially if there’s a requirement that vocals be generated using Synthesizer-V. At least then you know that some effort has been put in on the production, and it utilizes the product that brought everyone here in the first place.
I completely understand your hate for AI-generated art and music, Sparkelep, but creating a post like that isn’t gonna change anything. Synthesizer V is still AI—that’s literally what the ‘AI’ stands for in its name. If you’re gonna gatekeep AI usage, why stop at art and music? If you hate Ai Art/Music so much, then why did you join Dreamitonics Users Forum where Ai is literally half the conversation? If you don’t have anything nice to say, maybe don’t say anything at all—unless you wanna spend your forum days yelling at clouds.
You’re right! Synth V IS still AI, but unlike AI Art or AI Music there are two things that make Synth V different. Ethics and way of use. For something like Stable Diffusion or Suno you just press and button and you’re left with something that’s instant but lacks personal touch. It may have your lyrics and your input as the prompt but you didn’t actually make it 100%. For Synth V you have to work to get it to sound the way you want, either through the pitch tuning or breathiness or gender or loudness or even using the provided vocal modes. Synth V involves ethics and effort. AI Art and AI Music don’t. And the only reason I made the post is because it’s very grey as to whether Dreamtonics supports AI Art and AI Music since they sponsored someone in the past who seemed really into one of the two. If DT’s really the ethics based company they are then at the very least they need to be transparent about where they stand. If they’re ok with it then they should at least make sure to form rules that say “Hey you need to label this post as using AI or not or risk deletion”. But if they’re not ok with it then the next step’s to make sure it’s enforced and that they aren’t seen as an AI slop company. Synth V promotes hard work through working with ethically sourced voice databases but suno and stable diffusion don’t. Those two are soulless instant gratification tools and it’s really sad that Synth V gets lumped in with them sometimes just because they use AI too. I’m not gonna go on everyone’s posts and try to say it’s wrong but at least with this post I can be honest without going after anyone specifically. There are far better ways to get through the creative process that don’t involve using Suno or any image generators. Think about what people did before these AI things existed! They used royalty free song samples or used clip art or royalty free artwork. That’s ten times better than just putting a prompt in Suno and saying “Hey make a song like this”. I know I’m not gonna convince everyone ofc but Synth V means a lot to me and I don’t want it conflated to some ai slop machine that Suno and other stuff is. It’s way too special and different even though it does use ai.
When people refer to AI, they are generally referring to generative AI. Synthesizer-V is not generative AI.
Machine learning is used to create a data model of a voice, but then once the voice model has been created, it works in the same way as a non-AI sample library or physical modeling VI (or most likely a hybrid of the two).
I understand the backlash over generative AI, even though I don’t personally think its all bad. When @sparklep said “For something like Stable Diffusion or Suno you just press and button and you’re left with something that’s instant but lacks personal touch.”, that’s not the whole story, and unfortunately, this spouting as fact is where half the problem is. You can provide your own data to Suno for example, and it will use it. Music, lyrics, guitar solos… anything. That doesn’t address the moral issue of where they trained their models, but it is misleading to suggest that Suno is only good for AI slop.
This is completely unrealistic. Unethical ai users are a huge customer base for dreamtonics. Of course they are going to be supported. And frankly I’m glad that synthv has reached such a wide audience and had this success. It means DT has more resources to keep improving the engine.
Personally, I think there’s a place for all types of Synth V work here, but there should be a way to flag posts that contain generative AI. Alternatively, perhaps it could have its own section. Is that workable? I don’t know.
Sorry for confusing you.
Who cares? Especially since Synth V is synthetic.
AI tools and generative AI are very different categories.
More or less everyone using electricity is using some kind of AI Tool. An operating system is an AI Tool. A basic word processor is an AI Tool. Synthesizer V is an AI Tool. These tools don’t generate content – they don’t write lyrics or toplines, write prose, or create art.
Generative AI is very different. It does generate ideas and material with relatively little human input and effort. You’re not having the same human interaction and engagement as you would with legitimate art, and that is a very important distinction.
You zoomed in on one of the unspoken issues that matter. I agree that drawing a line between “AI songs” and using AI tools to enhance a composer’s original work are two different things.
I see this in a different light. If someone enjoys putting together something using AI and they want to label it as such and post it, more power to them. Should AI replace real composition and music jobs? Absolutely not. But if someone comes up with an idea and can’t create it themself they can use AI to produce what they have in their head. What’s wrong with that? If you see an AI label you can simply skip it, no harm done. Others may find the piece interesting, fun or inspirational. And that’s what music is all about.
FWIW
Looking at what’s been said so far I probably jumped the gun by saying “Hey none of this should be on the forum”. While I still really don’t like stuff like Stable Diffusion or Suno, I have to acknowledge that a lot of people who use Synth V also use those types of AI and me saying “None of that should be here” without taking into consideration the alternative of it simply being labeled as such was wrong of me and I should’ve considered that option before asking for such a big thing to happen. I’m sorry for jumping the gun and looking self righteous or morally superior in the process. Honestly it just being labeled as using Suno or AI art’s enough. If it means a lot to other people then I can’t stop them from doing it and trying to police it isn’t the answer.
I think you’re trying to push the wrong buttons. AI is here to stay, and a lot of it is not the unethical stuff you’re labelling it as. Moises, SoundID Voice AI, Tonalics, and many others use identified, paid artists for their products. Even Suno now has a legitimising deal with at least part of the industry (though legal action continues with other bodies). In most cases, you won’t be able to identify which AI products were used and banning all of them from this or any other music forum will just lead to the forum itself becoming moribund. Pretty much every professional studio is using AI tools - including generative AI - to inspire their work efficiently, and you’ll never know which tracks were. at least in part, developed using AI. Are you going to abandon the entire music scene because you can’t tell what’s wholly human-produced?
Just ignore those posts and creators who you don’t like. It’s simple.
I get that tbh. I may not like it but I can’t just police it and trying to get rid of it entirely. Sorry about that