Google Gemini can now create original music from text prompts
What's the story
Google has integrated its advanced AI model, Lyria 3, into the Gemini app to generate music from text prompts and images. The new feature allows users to create 30-second tracks by simply describing the mood or genre they want. Unlike basic text-to-audio tools, Lyria 3 is more sophisticated and can produce high-fidelity soundtracks based on uploaded photos or videos.
User experience
Lyria 3 is a lightweight creative companion
The Lyria 3 model comes from DeepMind, Google's research lab behind some of its most sophisticated AI systems. The tool is designed to be a lightweight, creative companion instead of a professional music replacement. Users can either type a prompt describing their desired genre or mood, upload an image or short video to guide the soundtrack, or start from a template track and remix it using prompts.
Feature rollout
Rollout and copyright considerations
The rollout of Lyria 3 has started on the desktop version of Gemini via the web, with mobile app support expected in the coming days. Google says it has been "very mindful of copyright and partner agreements" during training. If an artist is mentioned in a prompt, "Gemini will take this as broad creative inspiration and create a track that shares a similar style or mood."
Copyright considerations
Checks against existing content
Google has made it clear that Lyria 3 is "designed for original expression, not for mimicking existing artists." There are filters in place to check outputs against existing content. However, the company also acknowledges that their approach might not be foolproof and you can report content that may violate your rights or the rights of others.
User prompts
How to create music in Gemini?
In the Gemini app, users can access the new music feature from the "Tools" menu. You can describe a specific genre, mood, inside joke, or memory to create unique tracks with lyrics or instrumental audio. The end result is a 30-second track with custom cover art generated by Nano Banana. All tracks are embedded with SynthID watermark for verification purposes.
Language support
Age and language restrictions
Lyria 3 in the Gemini app is currently available for all users aged 18 and above, in multiple languages including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese. Google AI Plus, Pro, and Ultra subscribers will have higher limits. The company plans to expand language support and improve quality in the future to make this advanced music generation tool more accessible.