Google Drops Lyria 3: AI Music Generation Hits the Gemini App
Google's latest generative music model creates 30-second tracks from text or images, now live in beta across eight languages.
Google just shipped Lyria 3, its most advanced generative music model, directly into the Gemini app. The tool lets anyone cook up 30-second music tracks using nothing but text prompts or images.
The model launches in beta and supports eight languages out of the gate. Tracks come paired with AI-generated cover art featuring what Google calls "Nano Banana" visuals — a quirky touch for a music generator living inside a chatbot.
This is Google pushing Gemini further beyond text and code into creative territory. Rather than building a standalone music app, the company is embedding generation capabilities where users already live.
Thirty seconds isn't a full song. But it's enough for creators to prototype ideas, build loops, or just mess around with AI-powered composition without touching a DAW.