Media & Culture

Google’s AI music maker is coming to the Gemini app

DeepMind's model generates 30-second tracks from text, images, or videos directly in chat.

Deep Dive

Google and DeepMind launched Lyria 3, a new AI music generation model, directly into the Gemini app. The model creates 30-second instrumental or lyrical tracks from text descriptions, uploaded images, or video references. Available globally in 8 languages, it's designed for personal expression, not mimicking artists. This marks Lyria's first major consumer release, moving beyond its previous limited access on Google Cloud's Vertex AI platform.

Why It Matters

Democratizes AI music creation for consumers, integrating generative audio directly into a mainstream AI assistant for the first time.