A new way to express yourself: Gemini can now create music
TL;DR
Google integrates its music generation model Lyria 3 directly into the Gemini chatbot.
Key Points
- Users can create 30-second music tracks via text or image prompts.
- The feature is live now in the Gemini app.
- Lyria 3 is Google's dedicated music generation model, not a general-purpose system.
- The tool targets creatives without formal musical training.
Nauti's Take
Music was long considered the last bastion of human creativity – or so many liked to believe. With Lyria 3 baked directly into Gemini, the barrier to entry drops to zero.
For AI practitioners, the notable move here is Google doubling down on specialized models over generalists. The real question is how long before ad agencies start replacing their sound designers.
Context
Music generation is landing in a mainstream chatbot interface – no separate tool, no expertise required. This drastically lowers the barrier to entry and puts Google in direct competition with specialized services like Suno or Udio. The model architecture matters: Lyria 3 was explicitly trained for music, suggesting higher quality and control than generic multimodal approaches.
For content creators and indie developers, this could further squeeze the market for licensed stock music.
Sources