Audio · Google DeepMind

Lyria 3

Google DeepMind's structured music model

[ Overview ]

Google DeepMind's music generation model produces high-fidelity stereo tracks up to three minutes with control over intros, verses, choruses, and bridges, available via Vertex AI and the Gemini app.

[ Pros ]

Understands song structure, not just loops
High-fidelity stereo output
Easy access via Gemini app

[ Cons ]

Vertex AI pricing less transparent than flat plans
Shorter max length than some rivals

[ Tags ]

music generationGooglevocalsAPI