AI Generated / Assisted Music

Yeah, getting something that feels right/done is basically: lots of credits ))
My biggest gripe on occasion is mostly getting suno to use my backing tracks properly and not go too crazy on the key changes, like some desperate attempt to middle 8 a track just because it's in the formula ( grr )

So having more control over the vocals is great. Shared your song to a few people.
Those vocals sound even better on the nearfields ))
.
If you need help give me a shout, I haven't figured out the nuances, but definitely helped me a lot.

My setup is basically using the lyrics as additional descriptions. 1000 characters really isn't enough. But basically I use chapgpt to describe the musical tone I am looking for give it the lyrics and let it add the specific tags, per lyrical line.

It has been really useful for EDM songs, I can control exactly where I get vocoder results pitch, reverb, delays.

I needed to learn basically like a sound engineer, what all the effects does, it is a hella of fun playing around, once you know what they do you can do really cool effects, honestly, adding delays, reverb and pitch shifts and micro manage every little detail.

But yeah you need to delve deep with the new model on suno it has a lot more nuances and way more complex.

But the tracks is doing well for less than a hour, but usually dies off dramatically after that

Tracks.jpg

Will publish this song later. gave it more upbeat melody and a bit more style changes, the record scratches is a nice touch.
 
Yeah another one having fun with this this voice, I have nothing better tonight, might as well wallow in self pity and write about it lol

damn okay, uplugged is now my favorite. I asked specifically to do the door/gunshot mimic effect at the end. Yeah it stupidly emotionally charged I like the drawing breath gives it that human touch this voice works great with unplugged it has so much emotionally range dipping between low and high pretty much spot on.

Looking forward to see how I can push it further with more styles without moving away too much from the core theme and style.
next will be to see if minimal synths can add another layer, it is something common industrial, but I don't want it to be the main focus just to change to beat up, next would be to see if can settle the vocal tone to high or low as well.Lots of ideas here to try out.

 
Thought I neaten the persona a bit with album art and add a contrasting track with a circus vibe.

I think it is a nice contrasting track, this one was a single generation, the circus tune is a nice touch asked for something similar didn't expect to be right on nose.



I love the art really nice touch and grabs your attention right away, will later generate a themed one for each song.

64b07510-d5f4-4aed-88d7-9dab0a570f20.png
 
This one was a hard one to get just right, took a couple of tries, to get it just right. Had to tag the shyte out of it, to get the right atmosphere. I wanted something unsettling with a broken melody getting the laugh just right also a beatch.


I present Freaks
 
I’ve found it works really well to use different passes and styles to achieve the overall vibe you want to merge with, rather than relying on a single prompt or endless tweaking and layering in Suno Studio. This track is a cover of Chopin’s Funeral March. The first pass music used a custom cinematic music model I made, which added a dark fantasy, gothic instrumental to my uploaded sample, which came out great. The second pass was done in the base 5.5 model and layered in a cinematic, glitchy industrial horror music box feel. By balancing prompts, sliders, and other settings, it kept the key elements from both styles that I wanted to preserve.

 

Attachments

  • output.gif
    output.gif
    4.4 MB · Views: 0
I’ve found it works really well to use different passes and styles to achieve the overall vibe you want to merge with, rather than relying on a single prompt or endless tweaking and layering in Suno Studio. This track is a cover of Chopin’s Funeral March. The first pass music used a custom cinematic music model I made, which added a dark fantasy, gothic instrumental to my uploaded sample, which came out great. The second pass was done in the base 5.5 model and layered in a cinematic, glitchy industrial horror music box feel. By balancing prompts, sliders, and other settings, it kept the key elements from both styles that I wanted to preserve.


It is interesting see different approaches, I like experimenting and tweaking. Freaks for example I thought that, laughter would be a great addition with a female voice and off I went, the broken and circus like melody came afterwards as well, I have no problem spending a few credits on trying stuff.

Honestly for some songs it is rare to hit first time, it is just the nature of music production. That said chatgpt knows what I like and setup defaults for suno production, I only ever use chatgpt to setup melodies and fine tune styles, I found suno to be pretty weak in certain areas, and if you don't micromanage certain things suno fcks it up and just applies a general blanket generation cues on it.

My focus has been to get more vocal range and emotive responce, but each has their own creative engine and fun to see how people go about it :)

Suno likes safe averages, unless you tell it exactly to break the melody intentionally it will always apply defaults. For my self I find it fun to poke it with a stick and some time brute force it and go in dry, some times it works some times it doesn't.

My next experiments is going to be with 8th and 16th notes and eventually throw 4th and 8th and 16th notes in the same song, I have even messed with the standard question and resolution formula and intentionally have a melody that doesn't resolve.

This was too see if I can intetionally force a unresolved melody, did an okay job, the melody never settles, now imagine throwing in a mixture of 4th 8th and 16th notes in there. But I will definitely go with a single unresolved melody next time instead of many.

Ordered chaos if you will.
 
Okay this is a weird one. uses 1 - 2 and 1 - 2 - 3 with an occasional 4th, I don't need to explain why it is hard to have uneven notes, and why most songs stick to 4/4 or 8th and 16th. The track pattern doesn't settle, I chose it here specifically as it fits perfectly here.
Took a few tweaks to get it right, casual listener would feel the unease but, it is the above average that will notice the pattern differences., it catches you are attention.

 
Okay this is a weird one. uses 1 - 2 and 1 - 2 - 3 with an occasional 4th, I don't need to explain why it is hard to have uneven notes, and why most songs stick to 4/4 or 8th and 16th. The track pattern doesn't settle, I chose it here specifically as it fits perfectly here.
Took a few tweaks to get it right, casual listener would feel the unease but, it is the above average that will notice the pattern differences., it catches you are attention.

Sounds great))

Here is old song from back in the day, and I was wandering if suno can handle 6/8 and 7/8 time signature. If added to styles it seems to follow, but it could just be using the backing track.

 
Sounds great))

Here is old song from back in the day, and I was wandering if suno can handle 6/8 and 7/8 time signature. If added to styles it seems to follow, but it could just be using the backing track.

Sounds like a trip on acid lol

For the occasional song it fine. But it hard to pull off consistently, I applied it to both the drums and the guitar. It is actually quite subtle, but if you know what you looking for you can actually count it in the drum beat, it mostly 1 - 2 drum beat some 1-2-3 beat and the cheeky 4th, but the rest is pretty hard to hard to notice.

Anyways I am done with my other persona ( will get back to the persona with a new theme and such)

The new one is male dark pop with some soft grunge and sheogaze.

I called it 3 states deep, as in 3 states deep into grieving. I added a few new songs, but mostly contains a smattering of covers of previous songs that fit the theme and style of this and bringing them under one persona.

Was generally all over the show and was a bit messy to be honest, so I am looking to consolidate songs under a single persona.

Personally I am not a fan of bright and cutesy music, this is far less darker, but it has more texture than "if you happy and you know it clap your hands' songs.....

My favorite is Untold Alibis


and

Static arguments ( likely my fav of the lot and also a cover)

 
Last edited:
Yeah , so silly. But it’s good that people are starting to moan at the slightest signs of enshitification))
I think Suno was hoping that by removing sight of how many credits remain, people might burn through it quicker. Sneaky, but at least it's back with the toggle.
 
I think Suno was hoping that by removing sight of how many credits remain, people might burn through it quicker. Sneaky, but at least it's back with the toggle.
I don't actually mind too much, but worth keeping an eye on, I get extra 500- 1000 credits now and then. I was super creative last month and ended up getting more credits. I try to stick to about 100 credits a day some times less.

But I am currently in a creative cycle so I am using a bit more credits than usual, so will get an extra 1000, it is like R140 odd. Really don't mind pro plan is now slightly on the low side but 10k of premier is too much, I don't really focus too much on hooks, or adding video hooks as it costs too much. If in a slower creating cycle will try to use up credits with hooks.

10k is too much and I can't justify the price at triple of the pro plan at $30 most of it will go to waste, adding an extra 1000, is cheaper, I buy them in 500 credit batches, I don't use stems didn't use studio anyways. There is free editing tools if you really need to edit songs if you just use stems.

Personally it wouldn't surprise me if they up the credit cost of creating a song to 20 credits.

I read on their discord people complaining the relative bad exposure songs get and the static song lists, I listened to some of their lists and fck me there is some really **** music on it with 20k+- plays. There is definitely bias and "system gamed" songs on the lists.

What the people did was game the system with following every tom, dick and harry. I follow whoever follows me that way I get genuine reactions than "fake" ones and get fairer feedback on songs.

But really need to do a purge at some point consolidate them under a persona if it makes sense and delete the rest and make a cleaner list.
 
The problem with feedback and follows on Suno is that everyone is busy making stuff and they tend to believe their stuff is better than the next guy. I don't care too much over getting 500 likes, but gosh darn it it would be nice if once in a while someone stops and listens. I suppose the same is true on social media in general.
 
The problem with feedback and follows on Suno is that everyone is busy making stuff and they tend to believe their stuff is better than the next guy. I don't care too much over getting 500 likes, but gosh darn it it would be nice if once in a while someone stops and listens. I suppose the same is true on social media in general.
I am not looking for a billion views, but it would be nice to know you are heading in the right direction. System is honestly bugged.
But I discovered a lot of the people are tone deaf, they just publish any low effort song and then gets lost in the sea of decent songs.

I don't need someone blowing smoke, I would like feedback, but considering most aren't musically inclined, a song sung in the cave will sound to good to them.

So I had to teach my self and rely on my own ear it is less than ideal. So yeah I am considering casting a wider net and get a larger sample, x, facebook, tik rot.

I personally feel I have a few criminally underrated songs with low airplay.

That said youtube channel is slowly picking up it takes ages, majority of my demographics is in 35 - 45 age range. Which is great, it is demographic I consider with the best overall developed musical taste.

However I am also well aware AI music backlash and scrutiny which is why I haven't really pulled the trigger on distrokid yet or casting a wider net. I don't feel ready yet which is why I am experimenting and learning all I can, and why I am now focusing more on persona's (voices) and polishing the overall production quality. I don't want to half ass it.

Suno could be the perfect test ground, but it system is fckerer
 
I am not looking for a billion views, but it would be nice to know you are heading in the right direction. System is honestly bugged.
But I discovered a lot of the people are tone deaf, they just publish any low effort song and then gets lost in the sea of decent songs.

I don't need someone blowing smoke, I would like feedback, but considering most aren't musically inclined, a song sung in the cave will sound to good to them.

So I had to teach my self and rely on my own ear it is less than ideal. So yeah I am considering casting a wider net and get a larger sample, x, facebook, tik rot.

I personally feel I have a few criminally underrated songs with low airplay.

That said youtube channel is slowly picking up it takes ages, majority of my demographics is in 35 - 45 age range. Which is great, it is demographic I consider with the best overall developed musical taste.

However I am also well aware AI music backlash and scrutiny which is why I haven't really pulled the trigger on distrokid yet or casting a wider net. I don't feel ready yet which is why I am experimenting and learning all I can, and why I am now focusing more on persona's (voices) and polishing the overall production quality. I don't want to half ass it.

Suno could be the perfect test ground, but it system is fckerer
Yeah I get where you’re coming from.

I’ve already gone the distributor route to Spotify and the other platforms, so I can say from experience it’s a slow burn. It’s not like you upload and suddenly things move, it takes time to even get a trickle going. I’m sitting on a few hundred listeners a month now, which is cool, but it didn’t happen quickly.

YouTube is honestly the bigger grind for me. People think this stuff is low effort because “AI did it”, but they don’t see the actual work behind it. Making the track is one thing - which in itself is a considerable bit of work, then you’ve still got to do visuals, edit videos, upload properly, try market it, and somehow connect with people. It adds up quickly. Especially if you then need to push it to various other platforms as well.

For me it’s not really about chasing money or even big numbers. It’s more about putting stuff out there and sharing ideas, stories, whatever comes to mind with like-minded people, or people who enjoy the same style - or enjoy the humour, depending on the content. That part matters more than the stats.

But yeah… I do agree with you on one thing. Every now and then it would just be nice if someone actually stopped, listened properly, and said something real. Not hype, not noise, just genuine feedback. That’s the part that’s weirdly hard to get.
 
Yeah I get where you’re coming from.

I’ve already gone the distributor route to Spotify and the other platforms, so I can say from experience it’s a slow burn. It’s not like you upload and suddenly things move, it takes time to even get a trickle going. I’m sitting on a few hundred listeners a month now, which is cool, but it didn’t happen quickly.

YouTube is honestly the bigger grind for me. People think this stuff is low effort because “AI did it”, but they don’t see the actual work behind it. Making the track is one thing - which in itself is a considerable bit of work, then you’ve still got to do visuals, edit videos, upload properly, try market it, and somehow connect with people. It adds up quickly. Especially if you then need to push it to various other platforms as well.

For me it’s not really about chasing money or even big numbers. It’s more about putting stuff out there and sharing ideas, stories, whatever comes to mind with like-minded people, or people who enjoy the same style - or enjoy the humour, depending on the content. That part matters more than the stats.

But yeah… I do agree with you on one thing. Every now and then it would just be nice if someone actually stopped, listened properly, and said something real. Not hype, not noise, just genuine feedback. That’s the part that’s weirdly hard to get.
I am basically testing the waters to see what works, and not. I think for me is hitting that right voice. Since I have started getting technical with regards to voices, overall results are much better and more humanized.

I have been surrounded by music pretty much my entirely life, every genre you can imagine from the roaring 20's up to the late 2000's. So I think I have a decent ear with regards to commercial or radio quality songs.

Yeah I do this for fun and it relaxes the pants off me, it wouldn't hurt making a buck or two from it, however if I do, I do want to make sure I don't step on anyone's toes and be pretty straight forward that it is AI music and I make sure the lyrics at least the biggest contributing factor from my side, while I give direction on tone, doing specific instruments and move to technical specs and dictating exactly where why and when my final output is at least 40% - 50% human.

However I will say since version 4.5 it definitely became a lot more complex with regards to making music, you have waaaaaay more control in 5 and 5.5 really stepped it up with voice control and that you can micromanage it as well as do a lot more voice processed, even a little bit of tone and word inflections. I am finding unintentionally having to learn about things and experiment more.

I will do a couple more persona's and then pick the best one...
 
3 states pretty much down for now till I hit more inspiration

So onto the next one

27df8cae-c537-4565-8441-d86faa461b20.png
A bit of emo, alternative and ambient, I rarely ventured into emo as it can be be annoying and grind your ears. It is still on the darker side, very emotional and expressive, which is what I like. Experimenting with 4/4 and 8/8 notes and beats.

This time I went with generalized themed

My favorite so far is wreck me....the other two is a bit edgier, it wasn't the intention to head in that direction so kept it vague but easy enough to figure out the meaning. Yeah I am hitting the emo vibes a little on the nose, not sure if it a good or bad thing.

Leave a rose is also hauntingly beautiful, is a love letter to those left behind, just leave rose, don't mourn me basically.






 
c4283e05-9921-4d87-b2dd-2fc52c7ce880.png

Had absolutely nothing better to do today on a small break. So thought I might as well spend it "productively". Anyways did a second a "album", the previous one had a circus theme and very experiment. This one is a lot more raw, with grunge, and metal, industrial.The cover image is based on dante's inferno. This one isn't experiment just more raw and closer to grunge style with the addition metal. It is also a lot darker.


My favorites are stolen heart, pretty raw experience.

Shattered Hearts Club has a subtle breakbeat, something uncommon but it adds a nice texture.
 
Top
Sign up to the MyBroadband newsletter
X