Google is updating its Gemini app with a new music generation feature. At the moment, users can only generate 30-second tracks using a text prompt, but it’s also possible to create a unique track inspired by uploaded photos and videos.
“The goal of these tracks isn’t to create a musical masterpiece, but rather to give you a fun, unique way to express yourself,” Google explained today. Music generation in Gemini is powered by Lyria 3, the latest generative music model from Google’s DeepMind Labs. This model can also generate lyrics, and Gemini can create a unique cover art using Google’s Nano Banana model.
AI music generation has been popularized by apps like Suno, which can create professional-sounding tracks with lyrics and cover art for free. However, Suno requires users to pay a subscription to obtain commercial use rights for their AI-generated songs.
Google is being careful with music generation in Gemini, and the feature is currently in beta. All tracks generated in the app are embedded with a SynthID marker, and the company implemented safeguards to prevent its Lyria 3 model from plagiarizing existing content.
“Music generation with Lyria 3 is designed for original expression, not for mimicking existing artists. 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,” the company explained.
Music generation in Gemini is now rolling out for all users over 18 in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese. Google’s Lyria 3 model is available via Dream Track, a new YouTube Shorts creation tool that can generate a custom soundtrack based on a text prompt.