Google Lyria 3 Pro: AI Music Generator Now Creates 3-Minute Tracks & Integrates with Gemini

Google has unveiled Lyria 3 Pro, an enhanced version of its music generation AI model, initially announced in February. The upgraded model expands creative capabilities and broadens access to the technology, offering both enterprise solutions and integration within the Gemini application.

While the original Lyria 3 generated musical pieces up to 30 seconds in length, Lyria 3 Pro extends this to a maximum of three minutes. Crucially, the fresh model demonstrates an understanding of musical structure, allowing users to specify elements like intros, choruses, and bridges within their prompts. This feature enables the creation of diverse musical styles and complex compositions, according to Google.

The rollout of Lyria 3 Pro includes expanded integration with Google products. Gemini app users with paid subscriptions will gain access to the longer-form music generation capabilities. Google positions this as particularly useful for creating soundtracks for vlogs, podcasts, and tutorial videos.

Beyond Gemini, Lyria 3 Pro will likewise be available through Google AI Studio, the Gemini API, Vertex AI for large-scale production, Google Vids – a video creation application – and ProducerAI, a collaborative music production tool.

Google emphasizes its commitment to positioning these AI music generation tools as instruments for “creative expression.” The company states that Lyria 3 was developed based on insights gained from providing experimental tools to musicians, producers, and songwriters.

To address concerns regarding artistic originality, Google asserts that Lyria 3 and Gemini are designed to avoid imitating specific artists. Even when an artist’s name is included in a prompt, the system interprets it as “broad inspiration.” Generated music undergoes filtering to prevent copyright infringement, and an invisible watermark, “SynthID,” is embedded to identify the content as AI-generated.