ElevenLabs introduces a model capable of changing musical genres mid-track

ElevenLabs, known for its AI-based voice generation technologies, has launched its new music model called Music v2. The key feature of this model is its ability to seamlessly transition between musical genres within a single composition. According to company representatives, the new model can process both complex vocals and compositional structures with high precision. This is reported by Techcrunch.com .

ElevenLabs states that Music v2 can transition from opera to heavy metal, deliver fast-paced rap without losing logical consistency, and add non-musical sound effects to a track. Artists can also select specific parts of a song and use text prompts to regenerate them without affecting the rest. It is now possible to create sections like intros, verses, and choruses separately instead of just short clips, and then assemble them into a complete song.

In recent months, major laboratories such as Google, Stability AI, and Suno have been actively competing in the race for professional-grade music generation models. For example, at the Google I/O conference, features for creating covers and generating video clips via the Flow Music tool were demonstrated. However, ElevenLabs emphasizes that its model is built on licensed data and is fully safe for commercial use.

Such legal guarantees are crucial at a time when startups like Suno and Udio are facing copyright lawsuits. The new model is currently available for marketing teams via ElevenCreative and on the new ElevenMusic platform. Access for developers via ElevenAPI is expected to be available soon.