You've seen Gemini handle text, data, and images. But there's one thing that still catches people off guard — it can create original music too. Most people think making music requires talent, tools, or years of practice. With Gemini, all you need is an idea! Let's try it!

Behind Gemini's music creation is Lyria — the engine that turns your words into sound. You won't see it as a separate tool. It works quietly inside Gemini, handling everything: composition, instrumentation, vocals, and production. All you do is describe what you want — Lyria will figure out the rest.

Moreover, you don't need musical terms to do it well. You can say things like:

Lyria will build the musical interpretation from your words alone.

Thus, through effective prompting, you can create :

Let's see how it's done!

In the Gemini app, open the "Tools" menu and click the "Create Music" button — the one with the music note icon. That opens the music creation interface.

When you open Create Music, you'll see a set of ready-made styles like 90's rap, Folk ballad, Workout, and more. Each one is a starting point you can preview and customize. At the bottom, a prompt field lets you add your own instructions to shape the result.

Remix tracks are the fastest way to get started. Instead of building from zero, you:

The selected track sets the overall vibe and instrumentation. Your prompt then refines the mood, tempo, lyrics, or atmosphere.

And, here it is — a full track built from a base style, a mood, and a couple of instruments. Notice how the nostalgic feeling comes through in the melody and the guitar tone. That's Lyria reading between the lines of your prompt and the remix base.

Not bad for three parameters, right? Now imagine what happens when you give Gemini more to work with. The more layers you add, the more precise the result:

Let's see what happens when we keep the same track but change two things: tempo and vocals.

Hear the difference? Same instruments as before — but with a faster tempo and male vocals, the whole mood shifts.

So far, you've described music in words. But words aren't the only way in. You can also hand Gemini an image and let it do the listening — this is called Image-to-Audio. Gemini analyzes the colors, mood, and composition of a photo, then composes music that matches what it sees. This is especially useful at work. Upload a slide deck, a product photo, or any visual, and Gemini composes a matching soundtrack for it.

You can use Image-to-Audio for:

Remember the image of kids playing in the snow from the last lesson? Let's turn that into a track.

There it is — a full track generated from a single image. Gemini picked up the playful, energetic mood of kids in the snow and translated that directly into sound.

What do you do if the music is perfect, but the lyrics don't quite fit? You don't have to start over. Just ask Gemini to rewrite them while keeping everything else the same. Try: "Change the lyrics, but keep the rhythm, tone, and genre the same."

This is where the conversational advantage shines. Once you have a track, you can keep shaping it through chat:

Gemini will adjust each time without losing what's already working. This is like collaborating with a producer who instantly understands your vision.

Want to take it even further? You can change the entire song's language with a single request. For example, if you prompt something like "Change this song to French, while keeping the rhythm, tone, and genre the same", you can get this as a result!

You now know how Lyria works inside Gemini to turn your ideas into original music — whether you're describing a track in words, generating sound from images, or refining existing compositions through conversation. That's a complete toolkit for creating professional-quality audio without musical training!

With Gemini's creative tools now under your belt: images and music — you're ready for Unit 2, where you'll explore multi-step workflows and automations. See you there!