Pashto Medium

Voice AI company ElevenLabs has introduced ElevenLabs Music v2, an updated music-generation model that brings greater flexibility to AI-created songs. The release arrives nearly 10 months after the company launched the first version of its music generation technology.

The new model focuses on handling more advanced vocals and musical composition. One of its standout features allows songs to shift between genres within a single track. According to the company, users can move from opera to heavy metal and return without losing structure.

ElevenLabs Music v2 also supports fast-paced rap while maintaining coherence across vocals and arrangement. In addition, the model can insert non-musical sound effects directly into a composition.

Another update gives artists more control during editing. Users can select a specific section of a song and recreate it with text prompts. This process leaves the rest of the track unchanged, reducing the need to rebuild entire pieces.

The platform now supports building songs in sections instead of short clips. Creators can generate an intro, verse, and chorus separately, then combine them into a complete production. This workflow aims to improve control and speed during music creation.

ElevenLabs stated that the system performs more consistently across languages, lyrics, vocal styles, and arrangements. These improvements are designed to support broader creative use cases.

Competition in AI-generated music continues to grow. Companies such as Google, Stability AI, and Suno have also introduced music models capable of producing longer and more advanced tracks. During the Google I/O developer conference, Google added features for creating covers, editing songs by sections, and generating music videos through its Flow Music tool.

ElevenLabs highlighted another key point behind ElevenLabs Music v2. The model is trained on licensed data and approved for commercial use. Users can create tracks and use them without additional licensing concerns.

This approach has become increasingly important as some AI music startups, including Udio and Suno, have faced legal disputes related to copyright.

The model is currently available through ElevenCreative and the newly introduced ElevenMusic platform. Support for ElevenAPI is expected soon.