Google rolls out Lyria 3 to the Gemini app for creating 30-sec AI-generated music tracks

Google has integrated its most advanced generative music model, Lyria 3, into the Gemini app, introducing a new set of creative features. Users can now produce 30-second tracks by simply describing an idea or uploading an image to serve as inspiration. This update, currently available in beta, further streamlines music creation by generating lyrics automatically and allowing users to adjust style, vocal types, and tempo for greater creative flexibility. Lyria 3 also improves musical complexity and realism compared to previous versions in the Lyria series.

Following these music generation upgrades, each track created in the Gemini app comes with custom cover art generated by Nano Banana. This makes it easy for users to share their unique audio creations with friends. Google highlights that these features are intended for playful self-expression rather than professional music production.

In terms of rollout, Lyria 3 is now accessible to Gemini users in the United States and is expanding to YouTube creators internationally. Additionally, every track produced with the Gemini app contains SynthID, an imperceptible watermark to help identify content generated by Google’s artificial intelligence.