Google expands its commitment to music generated by artificial intelligence with Lyria 3 Pro, a version that extends the duration of compositions and gives more creative control to those who use it. This changes the landscape for producers, filmmakers and creators who need longer, customizable pieces of music in recent projects.
The company confirmed that Lyria 3 Pro emerges as an evolution of Lyria 3, the model integrated into Gemini in February, and that it now allows creating pieces of up to 3 minutes with a deeper understanding of musical structure. In practice, this means that users can request specific parts of a song—such as intros, verses, bridges, or choruses—and get more elaborate transitions between sections.
Beyond the temporal extension, Google highlights the greatest degree of creative control: the system responds to fine indications about style, instrumentation and arrangements, facilitating experiments with genres and changes of atmosphere within the same track. At the same time, the company ensures that it maintains technical safeguards to prevent the faithful reproduction of existing works.
What Lyria 3 Pro offers and how it differs
In practical terms, these are the main new features that the Pro version brings and how they compare to its predecessor:
| Feature | Lyria 3 | Lyria 3 Pro |
|---|---|---|
| Maximum duration | 30 seconds | Up to 3 minutes |
| Creative control | Basic track generation | Applications by sections, arrangements and transitions |
| Musical understanding | Initial Capacities | Model with deeper composition analysis |
| Origin identification | Yes — digital watermark | Yeah – SynthID on all tracks |
| Availability | Integrated into Gemini and associated tools | Vertex AI, AI Studio, Gemini (subscribers), and production tools |
Where it is available and examples of use
Google indicated that Lyria 3 Pro It can now be used within platforms such as Vertex AI y AI Studioin addition to integrating the functionality in environments designed for real-time creation and musical production. It will also be accessible in the Gemini app for users with a paid subscription.
- Producers and composers: rapid prototyping of longer structures and stylistic variations.
- Cinema and advertising: soundtracks with precise transitions and duration adapted to scenes.
- Content creators: exclusive jingles or themes with control over each section.
To underline its role in creative projects, Google collaborated with well-known artists: a Grammy-winning producer participated in the soundtrack of a DeepMind short film using Lyria, and DJ François K is working on a piece intended for publication soon.
Transparency and limits: what rules Google applies
A relevant measure for the sector is that all tracks generated with Lyria 3 and Lyria 3 Pro include a digital watermark, SynthIDdesigned to indicate that the audio was produced by AI and avoid confusion with recordings interpreted by humans.
Additionally, Google states that the model is not designed to imitate voices or faithfully reproduce specific artists: mentioning a creator’s name in the instruction works as a broad stylistic reference, not as a command to replicate their voice. The system incorporates filters to minimize the recreation of existing material.
These decisions matter today because AI-generated music is no longer just an experiment: its longer duration and creative control place it in larger commercial and cultural projects. Responsible use, traceability through SynthID and limitations to avoid direct imitations are key elements for its acceptance and future regulation.
As technology advances, musicians, production houses and platforms will have to adapt contracts, validation processes and rights policies to integrate these tools without compromising human authorship or transparency for the public.
Similar items
Diego Soler explores technological trends and digital innovations. You will enjoy practical tutorials and direct advice to better understand artificial intelligence, cybersecurity and new technologies.