๐ฐTechCrunch AIโขStalecollected in 16m
ElevenLabs music model enables mid-track genre switching
๐กNew granular control for AI audio generation allows for precise editing of complex musical structures.
โก 30-Second TL;DR
What Changed
Users can regenerate specific segments of a song
Why It Matters
Enhances creative control for AI music producers, allowing for more complex and dynamic compositions.
What To Do Next
Experiment with the new regeneration tool to refine specific transitions in your AI-generated compositions.
Who should care:Creators & Designers
Key Points
- โขUsers can regenerate specific segments of a song
- โขSupports mid-track genre switching capabilities
- โขMaintains consistency for the remainder of the audio track
๐ง Deep Insight
Web-grounded analysis with 10 cited sources.
๐ Enhanced Key Takeaways
- โขElevenLabs' Music v2 model is trained exclusively on licensed data, ensuring that all generated tracks are cleared for commercial use, which addresses a significant legal and ethical concern in the AI music industry.
- โขBeyond genre switching, the Music v2 model offers granular control, allowing users to define instrumentation, tempo, and structure, and can embed non-musical sound effects directly within a track.
- โขThe platform includes 'Music Finetunes,' a feature that enables users to fine-tune the ElevenLabs Music model on their own original, non-copyrighted audio to create a personalized and consistent sonic identity.
- โขElevenLabs provides its music generation capabilities across three distinct platforms: ElevenMusic for individual creators and remixing, ElevenCreative for brands requiring licensed music at scale, and ElevenAPI for developers to integrate custom music generation programmatically.
- โขThe Music v2 model supports full song generation, including AI-generated lyrics and vocals, in multiple languages such as English, Spanish, German, and Japanese.
๐ ๏ธ Technical Deep Dive
- The core of the new feature is the Music v2 model, a Text-to-Music model designed to generate studio-grade audio from natural language prompts.
- It utilizes hierarchical sequence-to-sequence architectures to learn associations between natural language descriptions and audio characteristics, influencing harmonic content, timbre, and arrangement.
- The model is capable of understanding both natural language and specific musical terminology for precise control over generation.
- It supports multilingual output for vocals, enabling generation in languages like English, Spanish, German, and Japanese.
- The 'Music Inpainting API' provides developers with fine-grained control, allowing them to modify specific sections of a track, extend or trim passages, change lyrics, create seamless loops, or transform the style and structure of a composition.
- Users can also access stem separation, a paid feature that splits songs into 2 (vocals and instrumental), 4 (vocals, drums, bass, others), or 6 stems for more detailed mixing.
- Generated audio is provided in high-fidelity MP3 (44.1kHz, 128-192kbps) and WAV formats, with track durations ranging from a minimum of 3 seconds to a maximum of 5 minutes.
- The 'Music Finetunes' feature allows for custom model training by uploading user-owned original audio, enabling personalized stylistic generation.
๐ฎ Future ImplicationsAI analysis grounded in cited sources
AI music generation will increasingly prioritize licensed data and artist monetization models.
ElevenLabs' emphasis on licensed data and partnerships with labels/publishers suggests a growing industry trend towards legally sound and artist-friendly AI music creation, differentiating from platforms facing copyright lawsuits.
The distinction between music creation, streaming, and remixing platforms will blur further.
ElevenLabs' ElevenMusic platform combines streaming, discovery, and AI-powered creation/remix tools, indicating a move towards integrated ecosystems where listeners become collaborators.
AI music will evolve beyond imitation to develop 'AI-native' musical behaviors and new macro-genres.
As AI music tools become more sophisticated, they are starting to generate unique aesthetic traits and 'impossible genre hybrids' that transcend traditional human-defined genres, leading to new forms of musical expression.
โณ Timeline
2022
ElevenLabs founded by Piotr Dฤ
bkowski and Mati Staniszewski.
2023-01
ElevenLabs publicly released its beta platform and secured a $2 million pre-seed round.
2024-01
ElevenLabs raised $80 million in Series B funding, achieving a $1.1 billion valuation, and announced new products including AI Dubbing Studio.
2025-08
Eleven Music was initially launched, enabling generation of studio-quality music from natural language prompts.
2025-12
Eleven Music received updates including the Music Inpainting API for granular editing and stem separation.
2026-05
ElevenLabs officially launched the ElevenMusic platform and the Music v2 model, emphasizing licensed data and integrated creation/streaming/remixing.
๐ Sources (10)
Factual claims are grounded in the sources below. Forward-looking analysis is AI-generated interpretation.
๐ฐ
Weekly AI Recap
Read this week's curated digest of top AI events โ
๐Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: TechCrunch AI โ