Google’s Gemini Now Composes Music with Lyria 3 AI
San Francisco, CA – Google has integrated artificial intelligence into its Gemini virtual assistant, enabling users to generate original music from simple text or image prompts. This new capability, powered by the company’s latest AI model, Lyria 3, marks a significant step in democratizing music creation and offers a glimpse into the evolving intersection of technology and artistic expression. The development arrives as AI-powered creative tools continue to gain traction, raising both excitement and questions about copyright and artistic ownership.
Lyria 3, described by Google as its “most advanced music generation model yet,” allows users to create 30-second tracks complete with instrumentals, vocals, and lyrics. The system is designed to understand musicality, from rhythm to arrangement, thanks to collaboration with music producers and musicians during its development, according to Google. This isn’t simply about stringing notes together; it’s about crafting cohesive musical pieces based on user input. The tool is currently available globally in English, Japanese, Korean, Hindi, Spanish, Portuguese, German, and French for users aged 18 and over, and is accessible through the Gemini app.
The potential applications are broad. Google suggests users could transform a simple email into a track, or students and educators could generate custom instrumentals for presentations and projects. Content creators, particularly those in the rapidly expanding world of online video, are also expected to find the tool valuable, quickly generating background music and sound effects without the require for expensive studio time or licensing fees. The ability to pair a photo or video with a prompt and receive a track with matching lyrics further enhances the creative possibilities.
How Lyria 3 Works and What It Offers
Lyria 3 operates by translating user prompts – whether text-based descriptions like “a smooth indie rock track for a sunset drive” or image inputs – into musical compositions. The Gemini app then generates a 30-second track, complete with cover art created by another Google AI model, Nano Banana. Users can experiment with different genres, moods, and eras to tailor the music to their specific needs. Google provides a gallery of templates to help users get started, offering pre-designed tracks that can be customized with individual details. The process is designed to be intuitive and accessible, requiring no prior musical knowledge or technical expertise.
The system’s capabilities extend beyond simple song creation. According to Google Workspace Updates, Lyria 3 can be used to create “funny jingles to lo-fi beats,” demonstrating its versatility. The tool’s ability to generate both music and lyrics opens up possibilities for personalized soundtracks for a wide range of occasions, from special events to social media content. The speed and ease of use are particularly appealing to content creators who need to quickly produce engaging audio for their projects. As Asha Kiran Kumar wrote in Analytics Insight, this new AI creation is “greatly favored by content creators, as it enables them to add thumping beats to tech reviews and the music and effects that set the vibe for travel vlogs.”
Addressing Concerns About Copyright and AI-Generated Content
The emergence of AI music generators like Lyria 3 has inevitably raised concerns about copyright infringement. A report by Billboard highlighted that Google has not yet specified whether Lyria 3 was trained on music protected by copyright. Google’s response, as reported by multiple sources, is that it uses music it “has the right to use, as outlined in our terms of service, partner agreements, and in accordance with the law.” However, this statement has not fully allayed concerns among artists and rights holders who fear their work may be used without proper compensation or attribution.
To address these concerns and provide transparency, Google has implemented a “SynthID” watermark on all tracks generated by Lyria 3. This virtual seal identifies the music as AI-generated, helping to distinguish it from human-created content. This move is consistent with broader industry efforts to develop methods for identifying AI-generated media, a crucial step in navigating the ethical and legal challenges posed by these technologies. The SynthID watermark aims to foster trust and accountability in the rapidly evolving landscape of AI-generated content.
The Broader Implications of AI Music Generation
Lyria 3 represents a significant advancement in AI-powered music creation, but it’s also part of a larger trend. AI is increasingly being used in various aspects of the music industry, from composition and production to mastering and distribution. This trend has the potential to disrupt traditional business models and create new opportunities for artists and creators. However, it also raises important questions about the value of human creativity and the future of the music industry.
The accessibility of tools like Lyria 3 could empower a new generation of musicians and content creators, allowing them to express themselves and share their work with a wider audience. However, it also raises concerns about the potential for saturation and the devaluation of musical skills. The debate over the ethical and economic implications of AI-generated music is likely to continue as these technologies become more sophisticated and widespread. The Gemini app and related in-app tools are controlled by Generative AI settings in the Workspace Admin console, allowing administrators to manage access and usage.
Getting Started with Lyria 3
For Google Workspace customers and users with personal Google accounts who are 18 or older, access to Lyria 3 is automatically granted with access to the Gemini app. To begin creating music, simply select “Create music” from the tools menu within the Gemini app. Administrators can control access to the Gemini app and its features, including Lyria 3, through the Generative AI settings in the Workspace Admin console. Detailed instructions and resources are available in the Google Workspace Admin Help Center and Google Help documentation. Google’s official blog post provides further details on the rollout and features of Lyria 3.
The rollout of Lyria 3 is currently underway, with full visibility expected within 1-3 days for both Rapid Release and Scheduled Release domains, starting February 18, 2026. This phased rollout allows Google to monitor performance and address any issues before making the feature available to all users. The company is committed to continuously improving Lyria 3 and expanding its capabilities based on user feedback and technological advancements.
As AI continues to reshape the creative landscape, tools like Lyria 3 offer a tantalizing glimpse into the future of music. While questions about copyright and artistic value remain, the potential for innovation and accessibility is undeniable. The next step in the evolution of AI music generation will likely involve further refinement of the technology, increased transparency regarding data usage, and ongoing dialogue between artists, developers, and policymakers.
Google plans to continue expanding the availability of Lyria 3 and adding new features in the coming months. Users can stay updated on the latest developments by following the Gemini app updates and visiting the Google AI blog. The company is also actively seeking feedback from users to improve the tool and ensure it meets the needs of a diverse range of creators.
Key Takeaways:
- Google’s Gemini app now features Lyria 3, an AI model capable of generating 30-second music tracks from text or image prompts.
- Lyria 3 creates music with vocals, lyrics, and instrumentals, offering a range of genres and customization options.
- Google has implemented a SynthID watermark to identify AI-generated music and address copyright concerns.
- The tool is available globally in several languages for users aged 18 and over, with administrative controls for Google Workspace.
What are your thoughts on AI-generated music? Share your opinions and experiences in the comments below. Don’t forget to share this article with your network to spark a conversation about the future of music and artificial intelligence.
Worth a look