Google introduced a new music generation feature powered by its latest artificial intelligence model, Lyria 3, expanding creative tools within its Gemini app.
The model, developed by Google DeepMind, is rolling out in beta and allows users to generate 30-second music tracks based on text prompts, photos or videos. Users can describe a genre, mood or personal memory, and the app will produce a custom track with lyrics or instrumental elements, along with AI-generated cover art.
See also: Gemini 3 Deep Think Upgrade Targets Frontier Scientific Problems
Google said Lyria 3 improves on previous versions in three key areas, including automatic lyric generation, greater control over style and tempo, and more realistic and complex compositions. “Our goal with music generation in the Gemini app is to help you add a fun, custom soundtrack to your daily life,” the company said in a statement.
The tracks are embedded with SynthID, Google’s digital watermarking technology designed to identify AI-generated content. The Gemini app is also expanding its verification tools to include audio files, enabling users to upload content and check whether it was created using Google’s AI systems.
See also: Gemini Live Gains New Expressive Dialogue Capabilities
Lyria 3 will also be available through YouTube’s Dream Track feature, initially in the United States and gradually expanding to creators in other countries. According to Google, the technology is designed for original expression rather than replicating specific artists, and includes filters to compare outputs against existing works. The company said users must comply with its terms of service and AI usage policies, which prohibit intellectual property and privacy violations.
Source: Google