Google Launches Lyria 3 Music Generation in Gemini App

Google Launches Lyria 3 Music Generation in Gemini App

Google has integrated its latest music generation model, Lyria 3, into the Gemini app, according to announcements from both the Google AI Blog and DeepMind.

According to the Google AI Blog, Lyria 3 allows users to “create custom, high-quality 30-second tracks from text and images.” DeepMind described Lyria 3 as “our most advanced music generation model,” stating it empowers “anyone to make 30-second tracks using text or images.”

The feature is now available in the Gemini app, making music creation accessible directly within Google’s AI assistant platform. Both sources emphasize the dual input capability—users can generate music either from text descriptions or by using images as creative prompts.

Both announcements characterized this development as “a new way to express yourself,” positioning the technology as a creative tool for general users rather than exclusively for musicians or audio professionals. The 30-second track length suggests the feature is designed for short-form content creation, potentially for social media, presentations, or personal projects.

Neither source provided details about the training data, availability by region, or any content moderation measures associated with the music generation feature.