∿ Music hackers. Algoraves. Inventing, playing. Neural synthesis. 24/7 ai death metal. Stable Audio team. Open models. Mischief @harmonai_org @artblocks
Meet Stable Audio 3.0, the open-weight model family built for artistic experimentation.
This is our open invitation to experiment with generative audio. We believe the best innovations are still waiting to be built.
The 4-1-1 on 3.0:
📣 You own your outputs, and can distribute and commercialize them under the Stability AI Community License (up to $1 million in revenue).
🎵 New and improved capabilities include variable-length generation up to six minutes, and full song composition on portable devices, no GPU required.
✅ Trained on a fully licensed dataset.
🎨 You can customize the models on your own library with support for LoRa training, which we’ve documented for the first time.
More on the models 👇
I made a looper where you can record midi or audio, and have magenta or stable audio respectively transform it with a prompt. Beatbox and turn that into drums? Check. Play some rough chords and get an orchestra? Check :) Had this app in my head for years!
yeah students protesting the college teaching songwriting/composition class using Suno.
Which is fair i get it.
Why would i pay insane tuition to a school to be taught to use an AI app that does it for me? And the app doesn't teach me anything about music nor coding? I'm here to actually fucking learn something and grow.
And there's lots of reasons to be critiquing mainstream commercial AI products, and the philosophy/marketing of their execs & influencers
But there's alsooooooo a super cool department at the school called Electronic Production and Design (EPD) where they teach you to be a master inventor of all kinds of computer music-- they teach you to be code, use open source, make your own DIY local ML models, break them, turn them into live performance instruments. And yes all this shit is hard as fuck & mad interesting. (Even half of freshman CS students drop out at my school)
I would happily advise the protest to be more poignant about what it's actually protesting
ok this is what happened:
there was a little daub of hummus on one of their music USBs and they said to wipe it off
then the palimpsest of misinterpretation simply escalated
@giorgiosancristoforo - Haiku generative album.
Haiku - Is an open work (Eco); the listener becomes the performer.
Every time you open the app, it creates 600 different samples, so it never repeats itself (10^16 possible patterns). Microstoria meets Oval!
meet ORAM: a little agentic multi-track looper for wild generative audio exploration, created at @sonic_field 🦋
prompt, record, or upload a sound. loop it, shape it with quick FX, then summon a new sound from whatever is playing.
powered by @StabilityAI's increbile Stable Audio 3, fully local ✿
Insane week of releases this week! Feels we’re entering into the next chapter of neural synthesis. Less “here’s the full song you requested” more ppl building custom instruments. Leading the way are scientist-engineers busily shaving off milliseconds with cleverer algo design.
Really proud to share something we’ve been working on for a while: Magenta RealTime 2 (MTR2), a live music model that is highly interactive (MIDI, audio, text, lots of parameters) and low-latency (~200ms end-to-end), and runs locally on a MacBook!