Google is integrating Lyria 3, DeepMind’s music-generation model, directly into Gemini. Yes, in addition to generating images and videos guided by your prompts, Gemini will now be able to generate music too.
The new Gemini integration is still in beta, but regardless, it is ready to pack a punch.
Music generation works similar to its visual counterparts. You simply describe what you’re looking for, and Gemini will parse through your prompt to craft a relevant result.
“Just describe an idea or upload a photo, like ‘a comical R&B slow jam about a sock finding their match,’ and in a matter of seconds, Gemini will translate it into a high-quality, catchy track,” wrote Google.
After images and videos, Gemini can now generate music too
Compared to older Lyria models, Lyria 3 improves on audio generation in three main ways:
- No need to provide your own lyrics! They’ll be generated for you based on your prompt.
- You have more creative control over elements like the style, vocals and tempo you want.
- You can create more realistic and musically complex tracks.
Additionally, what makes the model special is its multimodal approach. Not only can it generate tracks from text, it can also do so from photos and videos. So while you’ll be able to prompt Gemini with something like “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,” you’ll also be able to upload a photo or video and prompt Gemini to generate a track based on it.
It’s worth noting that Lyria 3 in Gemini is limited to 30-second tracks. It integrates with Nano Banana to generate custom cover art for the track, which essentially wraps a pretty bow around the whole experience.
Music generation is rolling out in the Gemini app for all users 18+ in English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese. It is currently limited to desktop, with support for the mobile app coming “over the next several days.”
