#ai-music-generation

#ai-music-generation

Spotify and Universal Music Group (UMG) have officially signed a licensing agreement that will allow "AI-enabled superfans" to produce AI covers and remixes of participating UMG songs.

Melo-D is a compact AI-powered instrument that combines generative music tools, guided learning systems, and a foldable guitar-like form factor into a single device. Developed by TemPolor, the project rethinks how users interact with musical instruments by replacing conventional music training methods with visual prompts, AI-assisted composition, and simplified chord systems.

The large model is available only through the API and self-hosting paid services. Plus, companies with more than $1 million in revenue would need to get an enterprise license. Many companies, including Google and ElevenLabs, are releasing models and tooling around music generation. However, as Suno and Udio's ongoing court battles have proved, licensing of data and partnerships with music labels could become a key part of the long-term survival of these services.

We tend to think AI music tools are just gimmicks for social media creators, or that they're limited to basic beats. But it's hard to dismiss them when companies like Google, Meta and Stability AI are pouring resources into generative audio models that can produce full compositions in seconds.

Back in 2008, Transport for London came up with a ruse to dispel antisocial behaviour: it piped classical music into supposedly problematic stations in the crime hotspots of south London. I think that was when I realised just how far the association of classical music with relaxing affect instead of real emotion had gone. Once an entire genre has become associated with relaxification, it's enough for you to hear the sound of an orchestra and think, This isn't for me.

During a demo, Henry Phipps, a Suno product manager, pointed to a song we had the model generate that included a flute-like synth with what sounded like a ping-pong delay effect on it: "I've never heard that before in previous models... what that says to me is that the model understands that this is an isolated sound that's being affected and needs to be reproduced faithfully in different parts of the stereo field."