Making a song is now easier than making toast
Google is adding a music-generation feature to the Gemini app, powered by DeepMind’s new Lyria 3 model.
Users can describe the kind of song they want, and Gemini will create a short track with lyrics and AI-made cover art.
You can also upload a photo or video, and the tool will generate music to match the mood.
Google says Lyria 3 produces more realistic and detailed music, with controls for things like tempo, vocals, and style.
The model is also coming to YouTube through Dream Track, a feature that helps creators make AI-generated music.
This tool is now expanding globally after being limited to the U.S.
Google says the system isn’t meant to copy artists directly, though prompts may still generate music with a similar style.
In brief:
Gemini can now generate music using DeepMind’s Lyria 3.
YouTube’s Dream Track is expanding worldwide.
Watermarks and detection tools are being added as AI music grows.
Artists are side-eyeing this hard
All tracks will include a SynthID watermark, and Gemini will also be able to check if a song was AI-generated.
The feature is rolling out worldwide to Gemini users aged 18+, with support for multiple languages.
AI music continues to divide opinion.
Platforms are exploring how to monetise it, while the music industry raises copyright concerns, and some companies face lawsuits over training data.
Finally, the country album I always dreamed of making can be a reality! - MV