ElevenLabs has released Music v2, an upgraded AI music generation model. It delivers better vocals, instrumentation, and arrangements across all genres. A single song can shift between opera and heavy metal, handle fast rap, and embed sound effects, all without losing coherence.
Another new addition is improved inpainting. Users can now regenerate specific sections of a song without touching the rest. Songs can be built section by section. Multilingual support got an upgrade too.
Music v2 powers three platforms: ElevenMusic for musicians, the ElevenAPI for developers, and ElevenCreative for brands and content teams. API pricing sits at $0.15 per minute, with a max length of five minutes and audio quality of 44.1 kHz at 128-192 kbps.
According to Elevenlabs, the model was trained only on licensed data, partly through a partnership with music company Believe. All generated tracks are cleared for commercial use starting with the Starter+ plan. Try it here.
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。