Lyria 3 Pro: Google lets you create tracks up to 3 minutes | Keryc
Google presents Lyria 3 Pro, a more powerful version of its music-generation model that expands what's possible: now you can create tracks up to 3 minutes long with greater control over structure — intro, verses, choruses and bridges — and complex transitions. Want a backing track for a podcast, the song for your vlog, or a soundtrack for a game? This is designed to give you that extra creative space.
What Lyria 3 Pro brings
Lyria 3 Pro improves the model's musical understanding: it recognizes composition and lets you request specific sections within the same generation. That means in a single prompt you can ask for an atmospheric intro, a verse with a strong groove, and a bridge with a key change, without stitching fragments together manually.
It also supports longer tracks (up to 3 minutes), which opens up possibilities to create full pieces or drafts that are very close to the final product. It's ideal for experimenting with styles, assembling complete ideas, or iterating on songs with more elaborate transitions.
If you've ever been frustrated cutting and pasting loops to fake a full song, this cuts down that work a lot.
Where you can use it
Google is integrating Lyria 3 Pro into several products aimed at different kinds of users: from developers and businesses to individual creators.
Vertex AI
Lyria 3 Pro is in public preview on Vertex AI for enterprises that need audio at scale. Think generating custom soundtracks for games, sound effects, or batches of tracks for media platforms without losing quality.
Google AI Studio and Gemini API
For developers, the model arrives in AI Studio and the Gemini API, offering greater musical awareness and structural coherence. That makes it easier to build creative tools, plugins, or integrations in editing and collaboration platforms.
Google Vids and Gemini app
In Vids, Google's AI-assisted video creation app, Lyria 3 and Lyria 3 Pro let you add music that fits the style of your projects. The rollout starts for Google Workspace customers and Google AI Pro & Ultra subscribers.
In the Gemini app, longer generations appear first for paid subscribers, giving you more room to experiment and refine musical ideas for vlogs, tutorials, or podcast intros.
ProducerAI
ProducerAI, the collaborative music-creation tool, integrates Lyria 3 Pro to offer an agent-assisted experience that helps musicians and producers iterate on complete songs. It's available globally for free and paid users.
Collaboration with industry and responsibility
Google develops these tools in dialogue with musicians, producers and the industry through the Music AI Sandbox. Artists like Grammy-winning producer Yung Spielburg and DJ/producer François K have used Lyria in creative and production processes.
The company says that the training and design of Lyria 3 respect materials that YouTube and Google are allowed to use under their terms, partner agreements and applicable law. Important:
Lyria 3 and Gemini do not imitate specific artists; if you name a creator, the model takes that reference as broad inspiration.
There are filters to compare outputs with existing content and policies that prohibit uses violating copyright or privacy.
All outputs from Lyria 3 and Lyria 3 Pro include an invisible watermark called SynthID to identify content generated by Google.
In short: more creative power, with mechanisms to protect original expression and limit abuse.
What this means for you
If you're a creator, you now have a tool that reduces friction in composition and lets you iterate on complete ideas without constant jumps between fragments. If you're a developer or business, you can embed high-fidelity music generation into products and workflows at scale.
Of course, AI doesn't replace artistic judgment; it speeds up routines, suggests variations and gives you a richer first draft. The best part? You can take bigger creative risks without spending hours on initial arrangements.
