Google Gemini App adds Lyria 3 AI music generation for text, photo and video prompts

Google has integrated AI music creation capabilities into the Gemini app with the introduction of Lyria 3, which expands its multimodal AI capabilities for creators. Google DeepMind introduced a new AI system which allows users to generate original music tracks through basic text inputs on Android and iOS devices.

The AI music creation capability is intended for social media short videos and digital storytelling, which serves as a major advancement to the Google Gemini application.

What can Lyria 3 do?

Lyria 3 enables users to create up to 30-second tracks by describing genre, mood, tempo or instruments. The AI music generator then produces a polished audio clip with layered instrumentation. Compared to earlier versions, Lyria 3 delivers improved sound quality, richer arrangements and better coherence across compositions.

Google DeepMind said the model can also generate lyrics automatically and offer greater control over style, vocals and tempo. Users can create music through two methods, which include their text prompts and their ability to work with visual elements. A photo of a sunset can be transformed into music which reflects the beauty of the sunset. Each music track can also have cover art created by AI.

Google is extending Lyria’s capabilities to YouTube Dream Track, allowing creators to experiment with AI-generated music for Shorts and other formats.