Google Rolls Out Lyria 3 AI Music in Gemini App

Google announced on Wednesday the rollout of Lyria 3, its latest AI music generation model, integrated into the Gemini app to enable users to create 30-second tracks from text prompts, photos, or videos. Developed by Google DeepMind, the feature marks a significant expansion of Gemini's multimodal capabilities, allowing global users aged 18 and older to produce custom music with lyrics, instrumentals, and accompanying cover art generated by the Nano Banana image model.

The tool operates through Gemini's interface, where users select the music option from the tools menu and describe a genre, mood, or concept, such as a nostalgic track about a childhood memory or a humorous jingle based on an uploaded photo of a pet. Lyria 3 automatically generates lyrics and allows adjustments to elements like style, vocals, tempo, and complexity, producing more realistic and layered audio compared to earlier versions. For instance, a prompt for a comical R&B slow jam about a lost sock could yield a full track ready for download or sharing via link. Google described the feature as designed for original expression, emphasizing its role in adding soundtracks to everyday moments rather than...