Google unveils Lyria 3 Pro with extended AI music generation capabilities
27 Mar 2026 - 14:38
Google has announced an upgraded version of its AI music generation system, introducing Lyria 3 Pro, a model designed to create longer and more structured tracks compared with its predecessor released only weeks earlier.
The original Lyria 3 model allows users to generate music clips of up to 30 seconds. The new Pro version extends this capability to tracks of up to three minutes and adds enhanced control over composition, including the ability to define elements such as intros, verses, choruses and bridges.
Google said the model has improved understanding of musical structure, enabling more detailed prompting and greater creative precision.
François K, a DJ and producer who worked with Google on the development process, highlighted improvements in sound quality and usability for professional workflows.
“Having explored the vast landscape of music tools, the progress on Lyria 3 is incredible, especially the fidelity and musicality,” he said. “The fashion in which I use generative AI tools never boils down to ‘one-button-click’ prompting. Instead, it’s becoming a versatile part of my arsenal, allowing me to refine ideas with realism and precision.”
Lyria 3 is currently available through Google’s Gemini AI application, while Lyria 3 Pro will be rolled out to the same platform but restricted to paid subscribers. The company also plans to integrate the Pro model into Google Vids and its AI music production tool ProducerAI, which it acquired recently. Additional integration is planned for enterprise products and the Gemini API.
Google said the system was trained using a combination of licensed datasets, partner contributions and permitted data from YouTube and Google services. It added that the model is not designed to replicate individual artists. However, it noted that when users reference artists in prompts, the system may draw “broad inspiration” from their styles.
All audio generated using Lyria 3 and Lyria 3 Pro will be embedded with SynthID, a digital watermark intended to indicate AI-generated content. The move comes as streaming platforms and digital service providers increasingly develop tools to detect and manage AI-generated music, amid concerns over synthetic or low-quality uploads flooding music catalogues.
Most popular
Call For Applications : ACCES seeks a UK-based performer
10 Sep 2026
Top amapiano songs of 2026
16 Feb 2026
Top Afrobeats songs of 2026
19 Mar 2026
Open call: Goethe-Institut grants for Southern African creatives
08 Sep 2026
AFRIMA 2026: All the winners
12 Jan 2026
Open call: Ernst von Siemens Music Foundation 2026 funding programmes
04 Sep 2026
Open call: FAME Week Africa, Pressure Cooker Studios songwriting camp
09 Sep 2026
Disclaimer: Music In Africa provides a platform for musicians and contributors to embed music and videos solely for promotional purposes. If any track or video embedded on this platform violates any copyrights please inform us immediately and we will take it down. Please read our Terms of Use for more.
Please log in to post a comment.