Google’s Gemini can now create 30-second music tracks for you
AI may not be advanced enough to handle your daily chores just yet, but creating music? That’s something it can do. Google has announced that Gemini can now generate 30-second music tracks based on simple text prompts. Just describe the kind of song you want, the mood or theme behind it, and Gemini will produce a short track within seconds. Interestingly, it can also create music using images as inspiration.
Google is using Lyria 3, DeepMind’s latest generative music model, to create music through Gemini. The easiest thing about creating music through Gemini is that you simply have to provide a prompt, and the AI will do the rest. So, you don’t need to have lyrics prepared, and you can also have more creative control over the style, vocals, and tempo. Google says you can “create more realistic and musically complex tracks” with Gemini.
Show Full Article
Here’s how it works.
There are two ways to create music with Gemini. One is through text prompts, where you can describe a specific genre, mood, or personal account to create a track. An example Google shared: “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.”
Another option is creating tracks from photos and videos. You can upload a photo or video and tell Gemini to compose based on that.
Gemini can create 30-second tracks based on your prompt and even generate a custom cover art with Nano Banana. You can then share the track by downloading it or simply clicking the share link. Google clarifies that these Gemini-created tracks aren’t supposed to be musical masterpieces but more of a “fun, unique way to express yourself.”
Google is also following strict protocols to make sure that there’s no copying of existing music by adding filters to check outputs against existing content. You can still name an artist, and Gemini will create the track in a similar style or mood. Google is also giving users the option to report any content that they think is violating their rights or those of others. All tracks generated by Gemini also have the SynthID watermark to clarify that it’s AI-generated content.
Lyria 3 has been launched in beta in the Gemini app and is available for all users above 18 in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese. It’s rolling out first to the desktop app and to the mobile app in the next few days. Google AI Plus, Pro, and Ultra subscribers will get to enjoy higher limits, so if you’re really enjoying this feature, then you can consider paying for these subscriptions.
Google’s Gemini is undoubtedly one of the most popular AI tools globally, but not the first to create music. There are multiple services available already, like MusicGPT, Suno, and Udio, that help you create full songs with any kind of beat or sound you want. Gemini’s approach is more fun and quicker compared to these platforms, and something you’d use to share with your friends and family.
