Stable Audio 3 Drops the Beat (and the Bass)!
“Jordan just hopes it doesn't try to auto-generate his stand-up routine.”
The story behind the toon
What happened: Stable Audio 3, an open-weight text-to-audio model, has been released. The Medium model generates up to 6 minutes 20 seconds of music and sound effects. Models run on NVIDIA GPUs (Medium) or CPUs (Small) and are available on Hugging Face with a GitHub repo. Outputs are royalty-free for personal and creative use under the Stability AI Community License.
This Neural Toon is based on the story “Announcing the release of Stable Audio 3!”.
Comic transcript
Panel 1 – Jordan: Another model release. Stable Audio 3.
Panel 1 – Taylor: OMG! Open-weight text-to-audio! So disruptive!
Panel 1 – Fact: Stable Audio 3, an open-weight text-to-audio model, has been released.
Panel 2 – Jordan: Music, SFX. Up to six minutes. Longer than expected.
Panel 2 – Taylor: Six minutes of AI jams! My podcast intros are saved!
Panel 2 – Fact: The Medium model generates up to 6 minutes 20 seconds of music and sound effects.
Panel 3 – Jordan: Runs on GPUs, even CPUs. Hugging Face, GitHub.
Panel 3 – Taylor: My laptop can run it! Democratizing audio generation!
Panel 3 – Fact: Models run on NVIDIA GPUs (Medium) or CPUs (Small) and are available on Hugging Face with a GitHub repo.
Panel 4 – Jordan: Royalty-free for personal use. Less licensing drama.
Panel 4 – Taylor: Unlimited, free audio! This is a paradigm shift!
Panel 4 – Fact: Outputs are royalty-free for personal and creative use under the Stability AI Community License.
More Neural Toons · Neural Snaps · The Neural Feed
📬 Get the top AI stories every morning
One email a day, no spam, unsubscribe anytime.