Gemini expands AI music generation capabilities

Google has introduced a significant upgrade to its Gemini platform, enabling users to generate full-length songs of up to three minutes through integration with its advanced Lyria 3 Pro music model, marking a notable step in the evolution of generative artificial intelligence for creative applications.

The enhancement positions Gemini as a more competitive tool in the fast-growing field of AI-generated music, where technology firms are racing to refine models capable of producing coherent, structured, and commercially viable compositions. The integration of Lyria 3 Pro builds on earlier iterations that were limited to shorter audio snippets, often criticised for lacking narrative flow and musical depth.

Google’s move reflects broader industry momentum towards longer-form generative outputs, with developers aiming to bridge the gap between experimental AI tools and practical use cases in entertainment, marketing, and independent music production. By extending track length to three minutes, the update aligns more closely with conventional song durations, opening the door to more realistic compositions that include verses, choruses, and instrumental transitions.

Executives involved in the project have emphasised improvements in coherence and stylistic consistency. Earlier models struggled with abrupt transitions or repetitive patterns, but Lyria 3 Pro is designed to maintain thematic continuity across longer sequences. This advancement is seen as critical in convincing users that AI-generated music can move beyond novelty and into mainstream creative workflows.

The upgrade also highlights intensifying competition among major technology firms investing in generative media. Companies such as OpenAI, Meta, and Stability AI have all introduced tools capable of producing audio, video, and images, each seeking to establish dominance in emerging creator ecosystems. Music generation, in particular, has become a focal point due to its commercial potential and relatively untapped market compared with text and image generation.

Analysts note that Google’s advantage lies in its integration strategy. By embedding Lyria 3 Pro within Gemini, the company reduces friction for users who already rely on its AI assistant for writing, coding, and research tasks. This convergence of capabilities could accelerate adoption, especially among content creators seeking all-in-one platforms.

At the same time, the rollout raises questions about intellectual property and ethical use. The music industry has expressed concern over how AI models are trained, particularly if copyrighted material is used without clear licensing frameworks. Legal scrutiny has intensified as record labels and rights holders seek greater transparency on data sources and compensation mechanisms.

Google has indicated that safeguards are in place to prevent misuse, including filters designed to avoid generating content that closely mimics identifiable artists or copyrighted works. However, enforcement remains a challenge across the industry, with regulators in multiple jurisdictions examining how generative AI intersects with existing copyright laws.

Beyond legal considerations, the technology is reshaping the economics of music production. Independent artists and small studios may benefit from lower barriers to entry, using AI tools to compose, arrange, and refine tracks without the need for extensive resources. This democratisation of music creation is widely viewed as a positive development, though it also introduces concerns about market saturation and the potential devaluation of human-created work.

Industry observers point to a shift in how music may be consumed and produced. AI-generated tracks could be tailored in real time for specific audiences, moods, or contexts, such as personalised playlists or adaptive soundtracks for games and virtual environments. This level of customisation, while technologically compelling, challenges traditional models of authorship and artistic identity.

Developers working on generative music systems argue that AI should be seen as a collaborative tool rather than a replacement for human creativity. By automating repetitive aspects of composition, the technology allows artists to focus on higher-level creative decisions. Critics, however, caution that widespread reliance on AI could lead to homogenisation, with algorithms favouring patterns that maximise engagement rather than originality.

The introduction of longer song generation also signals progress in underlying model architecture. Producing extended audio sequences requires greater computational efficiency and improved handling of temporal dependencies, areas where earlier systems faced limitations. Lyria 3 Pro’s performance suggests that these technical hurdles are gradually being addressed, paving the way for more sophisticated outputs.

Commercial implications are likely to emerge as the technology matures. Brands and advertisers are already exploring AI-generated music for campaigns, attracted by the ability to create bespoke tracks quickly and at lower cost. Streaming platforms may also experiment with integrating AI-generated content into their libraries, though such moves could provoke resistance from artists and industry groups.