On Thursday, 19 February, Google announced the launch of Lyria 3, the latest generative music model from Google DeepMind that will enable users to create personalized audio greetings and tracks in Arabic.
With Lyria 3 now available in beta on the Google Gemini website and shortly to be accessible on mobile, users can create original music tracks simply by describing what they wish to hear.
For example, one might request “an upbeat, modern Arabic fusion track for Ramadan,” and within seconds, Gemini transforms that prompt into a high-quality 30-second audio piece.
The model allows users to generate music in various ways, either through descriptive prompts or by uploading photos and videos to enable Gemini to compose soundtracks that reflect the uploaded content’s mood.
Each generated track comes with custom cover art produced by the NanoBanana tool which allows for easy sharing with friends via downloadable links or direct sharing options.
While Lyria 3 prioritizes creative expression, Google has implemented measures to safeguard the originality of the music produced. If a prompt mentions a specific artist, Gemini interprets this as broad inspiration, crafting tracks that capture a similar style while maintaining originality.
As part of this launch, users can also generate personalized Ramadan greeting cards using templates powered by the NanoBanana model. This feature enables the instant creation of custom cards, available in both English and Arabic, through a dedicated website.
Discussion about this post