ElevenLabs released Music v2 on May 28, a generative music model whose headline differentiator is the ability to switch genres in the middle of a track while preserving vocal identity, instrumental phrasing, and compositional structure. The release follows the original Music v1 from late 2025 and positions ElevenLabs as a direct competitor to Suno and Udio in the long-form music generation space.

The technical story underneath the mid-track switching capability matters. Most existing music generation models treat each genre as a separate conditioning signal that effectively restarts the generation pipeline when the prompt changes. A track that begins as a jazz ballad and shifts to electronic dance music tends, in those systems, to introduce an audible seam at the transition. Music v2 reportedly handles the shift as a single continuous generation, which means the model has learned conditional structure that lets it modulate stylistic features without losing the underlying coherence.

The use case ElevenLabs appears to be targeting is structured content production: film score, video soundtrack, and podcast bed music, where compositions need to track narrative arcs that change emotional register mid-piece. Suno and Udio have focused more on the standalone song format. ElevenLabs is making a different bet about what music generation gets used for in commercial workflows.

The competitive landscape has shifted meaningfully since Stable Audio 3.0 (which we covered on May 21) opened up commercially licensed open-weight music generation. Music v2 sits on the proprietary side of that line: ElevenLabs has not announced open weights or a non-commercial-use license tier. The product is API-first, which means the licensing posture and pricing structure ElevenLabs sets at launch will determine which segment of the production market adopts it.

For audio production teams currently evaluating Suno, Udio, and Stable Audio for client deliverables, ElevenLabs Music v2 is worth a focused test on workflows that involve narrative score or transition cues. The mid-track switching claim is the differentiator most worth verifying against your specific use case before committing to a pricing tier.

Published on the ElevenLabs blog on 2026-05-28.