Google expands Music AI Sandbox toolset with Lyria 2 Model for professional music creation

Google has expanded its Music AI Sandbox, a set of experimental AI-powered tools for professional music creation, with the integration of Lyria 2, the latest music generation model from Google DeepMind. Designed to generate high-fidelity, genre-spanning audio, the updated toolset introduces new capabilities for musicians, producers, and songwriters.

The update introduces Lyria 2, which generates 48kHz stereo audio across various musical genres, aiming for professional-grade outputs. A companion model, Lyria RealTime, supports interactive music creation and manipulation in real-time, mixing genres, blending styles and shaping audio moment by moment. New tools include 'Create' for generating music from text or lyrics, 'Extend' for continuing audio clips, and 'Edit' for transforming audio moods or styles.

The expansion follows Lyria’s initial limited enterprise preview. For now, access to Music AI Sandbox remains limited to U.S.-based creators, who can join through a waitlist, with the possibility of broader availability in the future.