Google is expanding its AI creative toolkit with Gemini music generation, introducing a new feature inside the Gemini app powered by DeepMind’s Lyria 3 model. The update, currently in beta, allows users to create original songs simply by describing what they want to hear.

The move strengthens Google’s push into generative AI for creators, positioning Gemini not just as a chatbot, but as a full-fledged creative assistant.

Gemini Music Generation Powered by Lyria 3

At the heart of Gemini music generation is Lyria 3, DeepMind’s latest music generation model. Google says the new version improves on earlier iterations by producing more realistic, layered, and complex tracks.

Using the feature is straightforward. Users type a prompt describing the song they want — for example, a playful R&B ballad about a sock finding its match — and Gemini generates a 30-second track complete with lyrics. The app also creates matching cover art using Nano Banana.

Beyond text prompts, Gemini music generation also supports multimedia inputs. Users can upload a photo or video, and the AI will compose a song that reflects the mood or theme of the media file. This adds a new dimension to AI-assisted creativity, blending visual and audio generation in one workflow.

Customization and Creative Control

Google says users can refine their creations by adjusting style, tempo, and vocal elements. This level of customization makes Gemini music generation more than a novelty feature — it becomes a practical tool for experimentation.

Importantly, Google has clarified that the system is designed for original expression. While users can reference an artist in a prompt, Gemini will only draw broad stylistic inspiration rather than directly mimic that artist.

According to Google, the model includes safeguards to check outputs against existing content. If an artist’s name is included in a prompt, the generated track will reflect a similar mood or genre, not a direct copy.

YouTube Dream Track Expands Globally

Alongside Gemini music generation in the app, Google is rolling out Lyria 3 to YouTube creators through Dream Track. Previously limited to creators in the United States, Dream Track is now expanding globally.

This integration allows creators to generate AI-powered music for their videos, reinforcing YouTube’s broader adoption of AI-generated tools.

SynthID Watermarking and AI Detection

Every song produced through Gemini music generation will include a SynthID watermark, Google’s system for identifying AI-generated content. The company is also adding detection capabilities within Gemini itself.

Users will be able to upload a track and ask Gemini whether it was created using AI. This transparency feature addresses growing concerns about the authenticity of AI-generated media.

The music generation feature is rolling out worldwide to Gemini users aged 18 and above. It supports multiple languages, including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese.

Mixed Reactions to AI-Generated Music

AI-generated music continues to spark debate across the industry. While platforms such as YouTube and Spotify are exploring monetization opportunities and partnerships with music labels, lawsuits over copyright and training data remain active.

Some platforms have introduced detection tools to label AI-generated tracks and prevent fraudulent streaming activity. As Gemini music generation becomes more widely available, these broader industry tensions are likely to intensify.

Google’s latest move underscores how generative AI is reshaping creative workflows — not just for developers and writers, but now for musicians and content creators worldwide.