ElevenLabs’s new music generation model can switch genres mid-track
ElevenLabs has launched Music v2, an updated music generation model capable of switching genres mid-track and creating full songs in sections. This new version offers enhanced control for artists and is built on licensed data, allowing for commercial use without copyright concerns.
ElevenLabs, a voice AI company, has released Music v2, an advanced music generation model. This new version, launched ten months after its predecessor, can seamlessly switch genres within a single track, handling complex vocals and compositions.
The model allows for genre transitions, such as from opera to heavy metal, maintains coherence in fast-paced rap, and integrates non-musical sound effects. Artists can now edit specific parts of a song using prompts without affecting other sections.
Musicians can also construct complete songs block by block, including intros, verses, and choruses, and then assemble them. ElevenLabs states that Music v2 performs consistently well across various languages, lyrics, vocals, and arrangements.
The release comes amidst a competitive environment where other AI labs like Google, Stability AI, and Suno are also developing advanced music generation tools. However, ElevenLabs emphasizes that its model uses licensed data, ensuring commercial viability and avoiding the copyright disputes faced by some competitors.
Music v2 is currently available through ElevenLabs' ElevenCreative tool for marketing teams and the ElevenMusic platform for AI-generated songs, with future integration into ElevenAPI.
Related articles
Build real agentic apps using CUGA: two dozen working examples on a lightweight harness
CUGA, IBM's open-source Agent Harness, simplifies building agentic applications by handling infrastructure, allowing developers to focus on tools and prompts. It offers pre-assembled components for planning, execution, and state management, significantly reducing development time. CUGA has topped agent benchmarks like AppWorld and WebArena.
OpenAI launches new initiative to help find and patch open source bugs
OpenAI has launched "Patch the Planet," a new initiative in partnership with cybersecurity firm Trail of Bits, to enhance the security of open-source projects. This program aims to assist maintainers in identifying and patching bugs, utilizing OpenAI's AI-powered security tools while reducing the burden on project teams.
PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters
Baidu has released PP-OCRv6, an advanced optical character recognition (OCR) model supporting 50 languages. Available on Hugging Face, this version significantly improves accuracy and efficiency across various parameter sizes, from 1.5 million to 34.5 million, marking a substantial leap in multilingual OCR technology.