The AI Music Studio Just Got a Lot More Flexible
ElevenLabs, the company best known for its AI voice cloning and text-to-speech tools, has expanded into music generation — and its latest model comes with a feature that audio producers will immediately appreciate: the ability to change genres partway through a track without rebuilding everything from scratch.
The new model allows users to regenerate a specific section of a song — a verse, a bridge, a chorus — while leaving the rest of the audio untouched. That kind of granular control has been notably absent from most AI music tools, where small changes typically mean starting over or hoping the model stays consistent across multiple generations.
What Makes This Different
Most AI music generators treat a song as a single output. You describe what you want, get a result, and if something feels off you iterate on the whole thing. ElevenLabs' approach borrows from how human producers actually work in a DAW: region by region, stem by stem.
The genre-switching capability is particularly interesting. A track can start as lo-fi hip-hop, shift into something with live-band energy midway through, then ease back into ambient territory — all without the jarring seams that typically plague AI-generated audio transitions. The model is apparently trained to maintain cohesion across these shifts, handling tempo, key, and mood transitions more gracefully than earlier generation tools.
This has obvious appeal for content creators who need music that evolves over the course of a video, podcast, or reel. It also opens up genuinely creative possibilities for musicians using AI as a composition partner rather than a replacement.
ElevenLabs' Expanding Footprint
The company launched its first music generation features in 2024 and has been iterating steadily since. It already dominates large chunks of the AI voice market — its voice cloning technology is widely used in podcasting, audiobooks, and video production — and music is a natural adjacent territory.
The move puts ElevenLabs in more direct competition with Suno, Udio, and Google's MusicFX, all of which have built sizable user bases among hobbyist creators and professional producers alike. What differentiates ElevenLabs' pitch is precision: the idea that AI music tools should give you the control of a proper editor, not just a prompt box.
The Bigger Picture for AI Audio
The pace of development in AI audio has been striking even by the standards of the broader AI boom. Eighteen months ago, AI-generated music was immediately identifiable — uncanny, flat, with suspicious drum patterns. Today's outputs are increasingly difficult to distinguish from produced tracks, and tools like this latest ElevenLabs model suggest the gap will keep closing.
For independent creators who can't afford session musicians or a full production budget, that's genuinely useful. For working musicians, it's a more complicated picture — the same tools that lower barriers to entry also lower the perceived value of production skill.
ElevenLabs hasn't announced pricing details for the new model's advanced features, though it's expected to be available through its existing subscription tiers.
Source: TechCrunch
