∿ Music hackers. Algoraves. Inventing, playing. Neural synthesis. 24/7 ai death metal. Stable Audio team. Open models. Mischief @harmonai_org @artblocks_io
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 👇
@priestessofdada yesss i love bytebeats as very minimal coding of generative music in 1 line of code
im imagining what the “1000 agents writing 600k lines of code” version is
maybe it could be based on bytebeats!
PyBot: No weights. No training.
A chatbot built with only simple rules and no AI, entirely written in 600k lines of Python code. PyBot matches Smollm with 360M parameters!
How? A swarm of 1,000 DeepSeek Flash agents wrote the whole code in just a week.
https://t.co/Du5GVvvLb5
Yay 🥳 We're TOP 14 flying to Bangkok 🇹🇭 next week for finals for this year's AI SONG CONTEST
Check out all the teams songs and VOTE for the winner here 🏆
https://t.co/S9hnyL035f
“….The best thing I've heard machine intelligence produce is still @dadabots excellent Calculating Calculating Infinity, released all the way back in 2017 with SampleRNN which is an ancient model.…”
With audio models (or in this case just... opus 5.5 with a guitar), it seems like every newer, larger model is a step backwards.
This particular qt'd example is just not something anyone would ever listen to, and I think part of it is that these models just end up producing something that sounds like nothing at all because of how large and generalized they are. Every new Lyria/Suno model seems to be trained in a way where the outputs are heavily weighted towards some generalized "average preference". So the fidelity increases, you can be more specific, and somehow it all still sounds completely uncompelling.
The best thing I've heard machine intelligence produce is still @dadabots excellent "Calculating Calculating Infinity", released all the way back in 2017 with SampleRNN which is an ancient model. Some of it is genuinely novel, retaining a distinct sound signature... and it's actually "hard" lol (check out this breakdown on Variations on a Cocktail Dressfinity from all the way to the end of the song, fucking nuts tbh)
Not saying that AI music won't ever be good, but I have a feeling that "good" AI music will come from a retvrn to smaller models trained on very specific source material.
just realized it is now technically possible to throw a party where every guest has a litemotif and the music in each room dynamically weaves everyone's themes together
my vocal synth models the vocal tract as a network of branching acoustic tubes, unlike vocaloid which stitches together and pitch-shifts phoneme recordings. still working on improving naturalness by tuning it against recordings of real singers, but I'm already impressed at how cool it sounds!
I manually inspected 40,000 animations to post train an LLM to make demoscenes.
Scraped 10 years of demos, did continued pretraining, curriculum learning and finally RLHF-ed with myself as the reward function.
Yes, models can have "taste".
All you need is data.
1/6
DADASPEC - I've been using this colorful perceptual spectrogram algo for all my players & demos since forever. It's really nice. It makes music structure rly obvious. Finally just named it & put it in a repo. Feel free to use. Give it to your claude.