Audiobus: Use your music apps together.
What is Audiobus? — Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.
Download on the App StoreAudiobus is the app that makes the rest of your setup better.
Your A.I. (Generative) Music+Sound Experiments
Most people who post and visit here are musicians and most of us have been making music for the majority of our lives (myself included). Having said that, I consider generative music services just another tool for lyricists, composers, producers and players. Most of all, they are just plain fun.
This thread is your invitation to post any fun or interesting experiments you may have made using "A.l." music tools to create original works, whether you contributed a little or a lot to the process.
If you like, describe what part you played in the final product. For example, if you input a prompt in the system and gave it a title or a particular feeling you were aiming for, tell us about it. If you took the output of a system and then re-recorded the whole thing to be more in your style, upload it and give some insight on your process.
Some sites where you can join (a number of them have free tiers) and start making music/sound experiments. If you have additional sites, list them:
Suno Al (best site I've found so far)
https://www.suno.ai
Loudly (pretty good for beats and short clips)
https://www.loudly.com
Soundful
https://my.soundful.com/signin
Google MusicFX
https://aitestkitchen.withgoogle.com/tools/music-fx
(Coming... Sometime Soon?)
Google Dream Track
Comments
Personally, I'm completely blown away with the sophistication I'm finding in the Suno AI system, which just released their version 3 software online. It's possible to generate up to 2 minutes of audio, but there is also a way to extend it beyond 2 minutes, even using the free tier.
On Suno AI I used the prompt "very slow baritone pop song" and even though it didn't deliver a baritone vocal, it really delivered on a piece of listenable music.
Lyrics the system generated (it should be noted that it is possible to supply your own lyrics to the system and then it will generate the rest):
[Verse]
In the shadows of my heart, I feel your presence near
Whispers of your love, softly spoken in my ear
But you're a thousand miles away, tearing me apart
In the darkness of the night, I hold onto your memory tight
[Verse 2]
Every moment, every breath, a bittersweet symphony
The melody of love, playing with such agony
I reach out for you, but you're just a distant star
Lost in the vast expanse, forever out of reach, so far
[Chorus]
Oh, I can't escape the longing, the ache down in my soul
These empty nights, this endless fight, I can't let go
In the depths of my despair, I'm drowning in the pain
But in the shadows of my heart, your love will still remain
This piece generated using Suno AI, I used the prompt "A sweeping, fast tempo orchestral score suggesting an adventure film" and let the system take it from there. It supplied the title and all music.
Title: "The Quest Begins"
Using Suno AI I provided the prompt "Uptempo pop-rock song about nervous laughter when faced with awkward situations" and it generated this title and music.
Title: "Awkward Grin"
By the way, for anyone who wants to experiment with Suno AI, here is their FAQ (with some tips on how to create a longer than 2 minute song among the most helpful bits of information).
https://suno-ai.notion.site/FAQs-b72601b96de44e5cacd2cd6baa985448
I don’t have any current audio examples to post exactly but I have been using AI a lot to help refine chord progressions a lot with fantastic results a lot of times. I’ve found it similar to how I use generative sequencers except that I get way more control to get exactly what I want out of it. It’s been useful for learning the theory as to why and how the chord progressions work as well.
I’ll have to try the audio stuff as well, I was using something a few months ago but lost interest. The results just keep getting better and better though, it’s becoming very helpful for the DIYers and more tangible to truly do everything yourself.
I don’t like AI.
I'm looking forward to prompting it creative weird stuff and getting creative weird stuff. Everything Suno gave me sounded like a hygiene product commercial or a tame John Wick score.
“Awkward Grin” surprised me. I suppose it won’t be too long before everything on Spotify is A.I.
I think for pop music and stock music it is inevitable. It’ll probably happen much faster than people imagine.
You will like A.I. ... whether you like it or not.
So Suno? Does it cost any money? If so, is it a tokens system, or subscription? (In this case I'd prefer the latter, lol.) See, when I produce something like a Justice track, I want to sample copyright-free tracks and recordings so that I may post them to Soundcloud.
Indeed, AI might be able to make some decent full tracks, but honestly it's simply sample fodder for me.
All of your questions are answered in their FAQ: https://suno-ai.notion.site/FAQs-b72601b96de44e5cacd2cd6baa985448
I could make six tracks for free. Three prompts, two tracks each. I think it would take a lot of re-rolls to hit on anything I would want to sample.
Hopefully some open source fine-tunable stuff comes out with tools that let you feed in your own audio etc.
Ooh, came back a couple hours later and had more credits.
Output (developer) also has one, email registration to use it. I messed around with it a bit but nothing that I wanted to use.
Output Co-Producer
Right. You get 50 free credits per day. Paid plans give you more.
All right, the tenth track hit hard AF, I'm in, lol
Excellent! Now I wasn't able to find this FAQ on their site, but I'm glad you found it for me mate.
Looks like a great deal to me. Up to 5 tracks per day is all I really need.
Hey, for the price of free it’s an incredible resource. And once you get used to using prompts to refine your requests, you might even consider the paid version. I’m considering it now.
Oh I did, I'm in! AI4LIF!
And if you sign up for a paid level you get the copyright on whatever you create (as long as you remain in their paid tier). You could release a couple of albums a month using it, easily.
I think this is the route I'll take since I don't want a copyright strike when sampling. 😂
I think ‘we’ need an A.I. (Generative) Music+Sound Experiments for Dummies video… totally lost..
Great for beatmakering for sure. Dont think I will ever release anything, but fun to mess around for sure.
I'm finding some help here (for Suno AI):
Suno.wiki:
https://www.suno.wiki
How to write a simple prompt:
https://www.suno.wiki/faq/getting-started/how-do-i-write-a-simple-prompt/
Song structure tags:
https://www.suno.wiki/faq/metatags/song-structure/
Musical style and genre list (these are the styles the system recognizes, so use these alone or in combination for prompts):
https://www.suno.wiki/faq/style-and-lyrics/styles-and-genres/
How to make your song longer:
https://www.suno.wiki/faq/making-music/how-do-i-make-the-song-longer/
And by the way, check out the formatting of this song.
https://app.suno.ai/song/d5c46ce9-bace-4fae-89f7-d28035e2cff7
Note the chord progression is shown in the lyrics. Brilliant. I'll have to try incorporating that into my experiments now.
And how about some Irish reggae? (No offense intended to anyone, this one was a test to see how genres would mix)...
Prompt: Reggae-inspired, synthpop, Irish comedy fable, uptempo
I'm posting this one just because it demonstrates what kind of very weird results are possible when you combine genres. I wrote out some quick lyrics and this one was very oddly interpreted by the system. It's so strange I think there's a germ of an idea here and maybe I'll develop it further later.
Prompt: Electropop, syncopated dance groove, funky vocals
Title: I Know What I Like
hm.. when i go to https://www.notion.so/suno-ai i end with this:
Nice thread. I was messing with MusicGen and StableAudio when they came out and even went so far and studied their training data (Shutterstock/Audiosparx) to get better ideas for prompts.
If you're interested in the extracted data (instruments, moods etc.) then I'll past some list into a post in this thread.
There's also Sunos Bark which works for speaker and announcer sample generation.
Didn't have the time yet to get into Sunos flagship product, but it's something I'll do sooner or later.
Around that time I've also decided to get deeper into electronic music and synths and this put the AI thing a bit on the back burner because I wasn't sampling that much.
Here are some short Hip Hop beat snippets I've created with MusicGen back then. They can be dropped and looped in a sampler like Koala where you set the right loop point, which was something I've explored for a while.