Google Lyria 3 Pro extends AI music generation from 30 seconds to 3 minutes

Google has expanded Lyria 3, its music-generation AI model, to create full three-minute tracks—six times longer than the previous 30-second maximum. The upgraded Lyria 3 Pro also introduces structural controls allowing users to specify song elements like intros, choruses, and bridges.

What's new

Lyria 3 Pro operates similarly to competitors Suno and Udio: users describe a mood, style, or instrumentation, and the model generates a corresponding track. The system can also create original lyrics from prompts or generate music based on reference photos or videos.

The length expansion addresses a critical gap for creators. While competitors already offered longer generation capabilities, Google's integration approach differentiates Lyria 3 Pro. The model is now accessible through:

• Gemini (no separate app required)
• Vertex AI (for enterprise customers)
• Google AI Studio and Gemini API (for developers)
• Google Vids (for office productivity)
• ProducerAI (Google's recently acquired music platform)

The ProducerAI integration is particularly significant, positioning Lyria 3 Pro as a direct competitor to Suno.

Copyright and authenticity safeguards

Google's expansion into full-length music generation invites immediate concerns about copyright infringement and artist impersonation. The company stated in its press release that "Lyria 3 and Gemini do not mimic artists," claiming the model treats artist names in prompts as "broad inspiration" rather than synthesis targets.

Google claims it performs output verification against existing content to prevent infringing material generation, and embeds a silent SynthID watermark in generated tracks to identify them as AI-produced. Whether these technical measures prove sufficient against determined misuse remains untested at scale.

What this means

The shift from 30-second clips to three-minute songs represents a maturation of commercial music AI, moving from novelty tool to potential production asset. For creators and enterprises, integration across Google's suite—particularly Gemini and Vertex AI—lowers barriers to adoption. The ProducerAI acquisition signals Google's commitment to competing directly with Suno in the music AI space.

However, full-song generation substantially amplifies copyright risks. Google's reliance on output filtering and watermarking is reactive rather than preventive. The music industry's response to whether these protections are adequate will likely determine regulatory pressure on the entire sector.