Rolling out in beta, Lyria 3, the latest generative music model from Google DeepMind, is now available inside the Gemini app, allowing users to turn simple prompts into original, shareable tracks within seconds.
Read More | Smart Rings or Smartwatches? What India’s Wearable Market Is Leaning Towards in 2026
Users can describe a concept in plain language or upload a photo or video as inspiration. A light-hearted prompt, a personal memory, or a visual moment can all become the foundation for a 30-second piece of music, complete with lyrics or instrumental arrangements. The aim is not studio-grade perfection, but a quick, creative way to give everyday moments a soundtrack.
This latest version represents a clear step forward from earlier Lyria models. Users no longer need to write their own lyrics; the system generates them based on the prompt. There is also greater control over musical elements such as tempo, vocal style and overall mood, resulting in tracks that feel richer and more musically detailed. Together, these upgrades make the experience more flexible while keeping it accessible to non-musicians.
The Process
The process works in two main ways. With text-to-track, users can describe a genre, emotion or personal anecdote, such as a nostalgic memory or an inside joke. Gemini will compose music to match that tone. Alternatively, images and videos can be uploaded and translated into music that reflects their atmosphere, turning visual memories into audio stories.
Each track is paired with custom cover art and can be easily shared via download or a simple link, making it easy to pass along to friends or family. It’s designed as a creative companion rather than a replacement for professional music tools.
Read More | Flexible, Low-Power, Intelligent: China’s Latest Chip Innovation for Wearables
Since first introducing Lyria in 2023, Google has developed the technology in close dialogue with the music community, including experiments such as Music AI Sandbox. That collaboration has shaped how Lyria 3 is trained and deployed, with careful attention to copyright, licensing and responsible use.
Originality over Imitation
Music created with Lyria 3 is intended to be original. When users reference well-known artists, Gemini treats those mentions as broad stylistic cues rather than instructions to imitate. Safeguards are in place to check outputs against existing works, and users can report any content they believe infringes on rights. Use of the feature is also governed by Google’s terms and generative AI policies, which prohibit violations of intellectual property and privacy.
Lyria 3 is available to users aged 18 and over in the Gemini app across several languages, including English, Hindi, Japanese, Korean, Spanish, French, German and Portuguese. The rollout begins on desktop currently. Subscribers to Google AI Plus, Pro and Ultra plans will receive higher usage limits.
