Google has introduced Lyria 3 Pro, an upgraded version of its AI music generation model that can create longer and more structured tracks. According to Google, the new model allows users to generate music up to three minutes long and is now being rolled out across multiple Google tools, including the Gemini app, Google Vids and more.
Lyria 3 Pro: What’s new
According to Google, Lyria 3 Pro builds on the earlier Lyria 3 model and focuses on creating more complete music pieces. The company said that the new model can better understand song structure, which means users can generate tracks with specific sections like intros, verses, choruses and bridges.
Google said that it has worked with musicians and industry professionals while developing its music tools. Through its Music AI Sandbox, the company has tested features with artists to improve creative workflows.
The company also highlighted that Lyria does not copy specific artists. If a user mentions a creator in a prompt, the system treats it as general inspiration rather than imitation. Outputs are also checked against existing content, and all generated audio includes SynthID watermarking to identify it as AI-generated.
Lyria 3 Pro: Availability
Google is expanding Lyria 3 Pro across several platforms to make AI music generation more accessible:
-
Vertex AI: The model is available in public preview for businesses that need large-scale audio generation, such as for games or creative platforms. -
Google AI Studio and Gemini API: Developers can use Lyria 3 Pro to build tools that generate structured music with better coherence. -
Google Vids: Users can add custom AI-generated music to videos. The rollout has started for Google Workspace users and AI Pro and Ultra subscribers. -
Gemini app: Paid users can generate longer tracks with more detailed prompts for personal or professional use. -
ProducerAI: A collaborative music tool that helps artists and creators refine full songs using AI. It is available globally for both free and paid users.
