2 Minutes
Ask Gemini for a song and it hands you a 30-second tune. Short. Designed. Ready to share.
Behind the scenes is Lyria 3, Google DeepMind's newest generative music model. As of today, music creation is rolling out in beta inside the Gemini app, available in English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese.
Anyone aged 18 or older can try the feature. Free users get access, while Google AI Plus, Pro and Ultra subscribers receive higher usage limits for heavier experimentation. To get started, type a prompt or upload a photo or video and tell Gemini what to riff on.
Cover art appears automatically courtesy of Nano Banana, and lyrics are generated for you — no need to write them yourself. Ask for a specific genre, a mood, an inside joke or a memory and the system will shape music and words to match that vibe.
Google frames these clips as playful tools for personal expression rather than attempts at full-length masterpieces. If you drop an artist name into your prompt, Gemini treats it as broad inspiration and aims for a similar style or mood instead of cloning a voice. Each track also carries SynthID, an imperceptible watermark meant to identify AI-created audio.
Safety measures are in place: outputs are checked against existing content, and users can report anything that might violate their rights or someone else’s. The result is a nimble way to prototype musical ideas, sketch sonic moods or just make something fun to share with friends.
Try a 30-second experiment and see what it sparks.
Source: gsmarena
Leave a Comment