ElevenLabs Releases Music v2, an AI Generator Capable of Genre-Switching AI voice company ElevenLabs is escalating the crowded race to build professional-grade music generators with the launch of its upgraded Music v2 model, promising more control for creators and a cleaner copyright story.

Launched nearly ten months after the company’s first music model, Music v2 is designed to handle more complex compositions and vocals. ElevenLabs says the new system can create songs that transition between styles mid-track, with the company highlighting that it can move “from opera to heavy metal” while maintaining coherence, even during faster rap delivery and when adding non-musical sound effects.

From a creator’s perspective, the update is as much about workflow as sound. The model lets artists regenerate only a section of a song using text prompts, without altering the rest of the track, and build full pieces by stitching together generated intros, verses, and choruses instead of relying on short, all-at-once clips. ElevenLabs says the system now performs more reliably across “languages, lyrics, vocals, and arrangements,” targeting marketers, branding teams, and independent musicians via its ElevenCreative tool and the new ElevenMusic platform, with API access promised soon.

Commercial use and licensing sit at the center of the launch. ElevenLabs emphasizes that Music v2 was trained “only on licensed data and cleared for commercial use,” positioning it as a safer option at a time when rivals like Suno and Udio have faced copyright lawsuits over their training sets. That stance aligns the company against a broader industry backdrop in which Google, Stability AI, Suno and others are rapidly rolling out tools capable of longer, more intricate tracks and even music-video generation.

Whether ElevenLabs’ licensing-focused approach will be enough to win over wary labels and artists remains to be seen, but Music v2 marks a clear attempt to distinguish its technology on both creative flexibility and legal assurances.

Continue reading https://foxvector.com