ElevenLabs's new music generation model can switch genres mid-track | TechCrunch

ElevenLabs launched Music v2, a music-generation model that can switch genres within a single track while handling vocal and compositional complexity. The model can transition between styles such as opera and heavy metal, produce fast rap with maintained coherence, and add non-musical sound effects. Artists can select a portion of a song and recreate it using prompts without altering other parts. Instead of generating only short clips, users can generate a song by sections like intro, verse, and chorus, then stitch the sections together. The model also performs more reliably across languages, lyrics, vocals, and arrangements. The model is built on licensed data cleared for commercial use and is available through ElevenCreative and ElevenMusic, with ElevenAPI planned.

"ElevenLabs noted that the model can go from opera to heavy metal and back, deliver fast rap without losing coherence, and can add non-musical sound effects to a track. With the new model, artists can pick a part of a song and recreate it using prompts without touching other parts of the track."

"Plus, instead of generating short clips, artists can build a song by sections, including the intro, verse, and chorus, then stitch them together. ElevenLabs added that the model performs more reliably across languages, lyrics, vocals, and arrangements."

"ElevenLabs emphasized that the new model is built on licensed data and cleared for commercial use, so users can freely use the tracks. Striking deals with labels is key, given that other AI music startups, like Suno and Udio, faced court cases over copyright issues."

"The new release comes nearly 10 months after the startup launched the first version of its music generation model. ElevenLabs launched a new version of its music-generation model, called Music v2, that can switch genres mid-track."

Read at TechCrunch