Google Gemini can now generate AI music tracks up to 30 seconds longAI image generators have flooded our feeds, text models are everywhere, and now Google is officially stepping into AI music creation. The tech giant is currently testing what it calls its “most advanced music generation model,” Lyria 3, rolling out in beta inside the Google Gemini app.
The update allows users to generate 30-second tracks from simple prompts, whether that’s a line of text, a photo, or even a short video clip. According to Google, Lyria 3 is a “high-fidelity AI music generator” capable of turning a funny moment, a specific feeling, or even a photo into a “custom soundtrack,” – complete with instrumentals, vocals and automatically generated lyrics.

READ MORE: Traxsource: “Any platform claiming foolproof AI detection is overstating what the technology can currently deliver”

The new model builds on Gemini’s existing ability to create text, images and video. It will also power the updated Dream Track feature in YouTube Shorts, enabling creators to generate backing tracks for short-form videos.
In practice, users can simply describe a genre, mood or memory and Gemini will compose a 30-second track to match, either instrumental or with lyrics. Prompts can be highly specific – for example: “I’m feeling nostalgic. Create a track for my mother about the great times we had as kids and the memories of her home cooked plantains. Make it a fun afrobeat track with a true African vibe.”
Alternatively, users can upload a photo or short video, which Gemini then analyses to generate music and lyrics that reflect the content and tone of the visual.
Google says Lyria 3 improves on its previous audio generation models by removing the need to supply your own lyrics, offering more detailed control over style, vocals and tempo, and producing more “realistic and musically complex” outputs. For now, tracks are capped at 30 seconds.
All music generated in Gemini is embedded with SynthID, Google’s imperceptible watermarking technology designed to identify AI-generated content.
Google is also clear about intent. “Music generation with Lyria 3 is designed for original expression, not for mimicking existing artists,” says the company. “If your prompt names a specific artist, Gemini will take this as broad creative inspiration and create a track that shares a similar style or mood. We also have filters in place to check outputs against existing content. We recognize that our approach might not be foolproof, so you can report content that may violate your rights or the rights of others.”

Learn more at Gemini.
The post Google Gemini can now generate AI music tracks up to 30 seconds long appeared first on MusicTech.