- What: Google DeepMind launched Lyria 3 Pro, an advanced music generation model with structural awareness.
- Capabilities: Supports tracks up to 3 minutes long with specific prompts for intros, verses, choruses, and bridges.
- Availability: Integrated across Vertex AI, Google AI Studio, Gemini App, Google Vids, and ProducerAI.
- Safety: All outputs are embedded with SynthID imperceptible watermarking to identify AI-generated content.
Google and Google DeepMind have significantly expanded their generative audio capabilities with the launch of Lyria 3 Pro, a new model capable of producing high-fidelity musical compositions up to three minutes in length. Announced on March 25, 2026, the update introduces "structural awareness," allowing creators to move beyond short clips to generate full-song architectures with distinct intros, verses, choruses, and bridges across Google’s enterprise and consumer ecosystems.
Moving Beyond Loops: The Power of Structural Awareness
The primary breakthrough of Lyria 3 Pro is its ability to understand the complex nuances of musical composition. While the standard Lyria 3 model—released just last month—was primarily focused on short-form tracks of approximately 30 seconds, the "Pro" iteration allows for extended durations that cater to podcasts, vlogs, and full-scale music production.
According to Myriam Hamed Torres, Senior Product Manager at Google DeepMind, the model's enhanced musical awareness allows users to prompt for specific transitions. This level of creative control means users can experiment with complex styles or generate songs that maintain coherence over several minutes, a significant technical hurdle in generative audio. The model does not merely extend audio; it understands where a chorus should peak or how a bridge should resolve back into a verse.
Enterprise and Developer Integration
Google is positioning Lyria 3 Pro as a versatile tool for professionals by integrating it into its core cloud and developer platforms:
- Vertex AI: Lyria 3 Pro has entered public preview on Vertex AI. This move targets businesses requiring on-demand audio at scale, such as gaming companies looking for bespoke soundtracks or video platforms needing high-fidelity background music.
- Google AI Studio and Gemini API: For developers, Lyria 3 Pro is now available alongside Lyria RealTime. This allows for the integration of structurally coherent music generation into third-party creative tools and applications.
- ProducerAI: This collaborative music creation tool, built by musicians, now offers an "agentic experience" powered by Lyria 3 Pro. It is designed to help artists and songwriters at all levels iterate on comprehensive song ideas globally.
Expanding to Consumer Creative Tools
The rollout also includes immediate updates to Google’s flagship productivity and AI apps. Google Vids, the AI-powered video creation app, now allows users to add custom music that matches the specific style of their marketing or creative projects. This feature is rolling out this week to Google Workspace customers and Google AI Pro and Ultra subscribers.
The Gemini app is also receiving the update, specifically for paid subscribers. Users can now use Lyria 3 Pro to create longer, more detailed tracks for personal projects like tutorials or social media content. This integration allows Gemini to transform text or images into custom soundtracks, providing a seamless bridge between visual and auditory creativity.
Responsible Development and Industry Partnership
Google emphasized that responsibility was foundational to the design and training of Lyria 3. The model was trained using materials that YouTube and Google have the rights to use under existing terms of service and partner agreements. To protect the rights of original creators, Lyria 3 and Gemini do not mimic specific artists; if a prompt names a creator, the model uses it only as "broad inspiration."
To ensure the technology serves human creativity, Google has been working through its Music AI Sandbox. This experimental suite has been used by Grammy-winning producer Yung Spielburg for the score of the Google DeepMind short film "Dear Upstairs Neighbors," and by legendary DJ and producer François K for a soon-to-be-released track.
"We have been developing our music generation tools responsibly and in close partnership with the industry to ensure AI serves as a tool for creative expression," Torres stated in the announcement.
Impact on the Industry
The shift from 30-second clips to 3-minute structured tracks marks a turning point for AI in the music industry. For the first time, AI music generation is moving from being a "novelty loop generator" to a legitimate production assistant capable of handling song-length arrangements.
For developers and businesses, this reduces the friction of sourcing licensed music for long-form content. For musicians, it offers a high-speed "sketchpad" for song structures.
"Lyria 3 Pro turns a text prompt into a full-length musical architecture, fundamentally changing the speed at which creators can move from an idea to a finished arrangement."
What’s Next
Lyria 3 Pro is currently rolling out to professionals, developers, and organizations. While enterprise features are in public preview on Vertex AI, consumer-facing features in Google Vids and the Gemini app are beginning their rollout to paid subscribers this week. Google has signaled that it will continue to expand the quality and language coverage of these models, with plans to bring more capabilities to both desktop and mobile platforms in the coming months.