Google has officially added generative music capabilities to its Gemini AI app through the integration of DeepMind’s newest model, Lyria 3, letting users create original 30-second music tracks from text descriptions, photos, or video uploads. The feature, currently in beta, produces custom songs complete with vocals, lyrics, and even bespoke cover art generated by Google’s Nano Banana model, all without requiring prior musical expertise. Tracks are watermarked with Google’s SynthID to identify them as AI-created, and the tool also offers creative controls such as tempo, style, and vocal options while operating under safeguards to avoid direct mimicry of specific artists. This rollout is available globally to users over the age of 18 in multiple major languages, and Google is expanding related tools like Dream Track for YouTube creators alongside the broader Gemini rollout.
Sources
https://techcrunch.com/2026/02/18/google-adds-music-generation-capabilities-to-the-gemini-app/
https://blog.google/innovation-and-ai/products/gemini-app/lyria-3/
https://gemini.google/overview/music-generation/
Key Takeaways
• Google’s Gemini app now includes AI-based music generation using the DeepMind Lyria 3 model, producing 30-second tracks from prompts including text, images, or video.
• The music feature offers automatic lyrics, creative controls (style, tempo, vocals), and SynthID watermarking to distinguish AI-generated content, with global availability in multiple languages for users aged 18 and older.
• Google is expanding distribution through related tools like Dream Track for YouTube and embedding it within the existing Gemini suite alongside other multimodal capabilities.
In-Depth
Google’s latest expansion of its Gemini app signals a broader push into multimodal creative AI by integrating Lyria 3, a refined music-generation model developed by Google DeepMind. This new feature elevates user access to generative AI by enabling everyday individuals—regardless of musical training—to create custom 30-second songs simply by typing a prompt or uploading an image or video that captures the mood they want translated into sound. For instance, telling Gemini to “create a laid-back pop groove for an evening barbecue” produces a fully composed track with instrumental backing, vocals, and on-the-fly lyrics that match the description. Cover art, created by Google’s Nano Banana image model, accompanies the song, making it ready for sharing with friends or on social media through direct download or link.
The inclusion of SynthID watermarking on each track represents Google’s effort to both promote transparency and offer a safeguard in the era of AI-generated media. This invisible mark helps identify content that originates from AI rather than human composition, an increasingly important feature as the regulatory and creative communities grapple with how to handle AI’s role in arts. Users also have a degree of creative control over the generated work, allowing selection of genre, tempo, and vocal attributes, which enhances personalization while keeping the tool intuitive for non-experts.
Globally available in eight major languages, the rollout targets adults aged 18 and older and is currently in beta, reflecting Google’s cautious approach to refining the experience before a full commercial launch. By adding this music-creation capability directly into the Gemini chat interface, Google reduces friction compared with standalone music-AI tools, situating generative audio creation alongside text, image, and video generation in one platform. This approach not only broadens the appeal of generative AI apps but also positions Google competitively against other tech firms that are introducing similar tools.
The feature’s release comes at a moment of heightened focus on AI and creative industries, where debates about copyright, artistic ownership, and economic impact are intensifying. While some stakeholders express concern that AI music could disrupt traditional artist revenue streams, Google emphasizes that Lyria 3 is designed to produce original compositions rather than replicate existing works. The introduction of AI-driven tools like Lyria 3 and their integration into widely used consumer apps like Gemini underlines the rapid evolution of generative AI from niche research to mainstream creative utility, reshaping how people make and interact with digital art.