Record scratch—Google’s Lyria 3 AI music model is coming to Gemini today (arstechnica.com)

Google has announced the integration of its Lyria 3 AI model into the Gemini app, allowing users to generate music without requiring lyrics. This marks a significant advancement in AI music generation, making it more accessible to a wider audience. With the new "Create music" feature, users can describe their desired sound or even upload an image to guide the mood, resulting in a 30-second track created in just seconds. This shift emphasizes the potential of AI to democratize artistic creation, challenging traditional notions of music as a solely human endeavor. Lyria 3's enhancements include faster processing and the ability to autonomously generate lyrics based on vague prompts, increasing the model's usability for novice users. Each generated piece will also come with a custom album cover created using the Nano Banana model, adding a visual element to the music. These tools will extend to Google’s Dream Track for YouTube Shorts, facilitating content creation across platforms. This innovation not only highlights the growing capabilities of AI in creative fields but also raises questions about the nature of authorship and artistry in an age of advanced machine learning.