Lyria 3 Pro: Create longer tracks in more

Google DeepMind has released Lyria 3 Pro, an advanced version of its music generation model capable of producing tracks up to three minutes long with distinct musical structures. The new model understands prompts for specific song elements such as intros, verses, and choruses, moving beyond simple clip generation toward more complete compositions. The announcement is significant as Google simultaneously begins a broad rollout of its Lyria models across its product ecosystem, targeting developers, enterprise customers, and general consumers.

The integration strategy spans multiple Google platforms. For businesses, Lyria 3 Pro is now available in public preview on Vertex AI, enabling scalable audio production for applications in gaming or creative tools. Developers can access the model through Google AI Studio and the Gemini API. The technology is also being embedded into consumer-facing products; Google Vids will use Lyria for custom soundtracks, and the Gemini app will offer longer track generation for its paid subscribers. Additionally, ProducerAI, a dedicated music creation tool from Google, will incorporate Lyria 3 Pro to assist artists and producers.

By deploying its music AI across its entire stack—from cloud infrastructure to consumer apps—Google is positioning its generative models as a foundational utility for content creation. The company also continues to address industry concerns by embedding SynthID watermarks in all outputs and implementing policies to prevent the direct mimicking of specific artists. This multi-pronged approach, which includes collaboration with musicians through its Music AI Sandbox, aims to integrate AI into creative workflows while navigating the complex intellectual property landscape of the music industry.

Google's strategy is not to build a single hit music app but to productize its foundational model by embedding it as a feature across its entire ecosystem. By integrating Lyria into Vertex AI, Gemini, and Google Vids, the company aims to make its AI an indispensable utility for a wide range of creators, solidifying its platform's value rather than competing on a single application.