Google just turned Gemini into a full music studio
Google's Lyria 3 Pro lets you create 3-minute songs with vocals in any genre — directly inside Gemini. Free tier included.
Google just upgraded its AI music generator from a 30-second clip maker to something that writes complete 3-minute songs — with vocals, lyrics, and custom cover art. It's called Lyria 3 Pro, and it works directly inside the Gemini app.
Type a sentence like "upbeat funk track with female vocals about a road trip" and Gemini will produce a full song with an intro, verses, chorus, and bridge. It understands song structure — not just mood — so you can shape the output the way an actual producer would.
30 Seconds Was a Toy — 3 Minutes Changes Everything
Last month Google launched Lyria 3, which could only generate 30-second clips. Fine for a ringtone — useless for a YouTube video, podcast intro, or presentation soundtrack.
Lyria 3 Pro fixes that. Here's what's new:
Lyria 3 Pro upgrades:
- 3-minute tracks — up from 30 seconds (6x longer)
- Song structure control — specify intros, verses, choruses, bridges separately
- Multi-language vocals — sing in multiple languages across pop, funk, Motown, jazz, and more
- Custom cover art — generated automatically by Google's Nano Banana model
- Export as MP3 or MP4 — ready for your projects immediately
You can also upload a photo and have Lyria 3 Pro write music inspired by the image. A sunset photo becomes an ambient track. A party photo becomes upbeat dance music.
How Much It Costs (and What You Get Free)
The basic Lyria 3 is free — you can generate 30-second tracks at no cost, right inside Gemini. For the full 3-minute Pro version, you need a paid Google subscription:
For context, competitors like Suno and Udio charge standalone subscriptions. Lyria 3 Pro comes bundled with your existing Google AI plan — meaning you might already have access.
Not Just the Gemini App
Lyria 3 Pro is rolling out across six Google products:
- Gemini app — the main consumer interface, available now
- Google Vids — for Workspace customers who need video soundtracks
- ProducerAI — Google's collaborative music creation platform
- Vertex AI — for enterprise-scale audio production
- Gemini API + AI Studio — for developers building music into their own apps
If you're a marketer making video ads, a podcaster who needs intros, or a YouTuber tired of royalty-free libraries — this is built for you.
50,000 AI Songs Hit Spotify Every Day
Here's the bigger picture. Roughly 50,000 AI-generated tracks are uploaded to Spotify daily. Last year, Spotify deleted 75 million tracks it flagged as spam. AI music is already flooding streaming platforms — and Google just gave it a major upgrade.
But there's a key difference: while competitors Suno and Udio are facing copyright lawsuits over their training data, Google says Lyria 3 Pro was trained exclusively on licensed and permissible content. Google also partnered with artists like Wyclef Jean during development.
Every track generated by Lyria 3 Pro includes a SynthID watermark (an invisible marker that identifies it as AI-made). You can even verify any audio file by uploading it to Gemini to check if it was AI-generated.
Try It Right Now
If you have the Gemini app (free or paid), you can start immediately:
Step 1: Open the Gemini app
Step 2: Tap Tools > Create music
Step 3: Describe what you want — genre, mood, tempo, vocal style, instruments
Step 4: Download as MP3 or MP4
Pro tip: Be specific. Instead of "happy song," try "upbeat acoustic folk with male vocals, 120 BPM, about a summer morning — start with a fingerpicked guitar intro, build into a full band chorus." The more detail you give, the better the output.
Free users get 30-second tracks (still useful for intros and short clips). Paid subscribers get the full 3-minute experience.
Related Content — Get Started with Easy Claude Code | Free Learning Guides | More AI News
Stay updated on AI news
Simple explanations of the latest AI developments