Google is extending its generative artificial intelligence ambitions into music, adding a new feature to its Gemini app that allows users to create 30 second songs from text prompts or uploaded images. The tool is powered by Lyria 3, the latest music generation model from Google DeepMind, and represents the company’s most direct effort yet to position artificial intelligence not just as a productivity aid, but as a medium for everyday creative expression.
Since the debut of the Gemini app, Google has emphasized visual creativity, enabling users to generate images and video. The addition of music suggests a broader strategic goal: to turn the app into a multimodal creative studio in which text, visuals and now audio can be produced on demand. Users can describe a concept in natural language or upload a photo and receive a fully formed track in seconds, complete with lyrics and cover art.
In announcing the rollout, Joël Yawili, Senior Product Manager for the Gemini app, and Myriam Hamed Torres, Senior Product Manager at Google DeepMind, framed the feature as an extension of the company’s creative toolkit. “Since launching the Gemini app, we’ve built tools to encourage creative expression through images and video. Today, we’re taking the next step: custom music generation,” they wrote. Lyria 3, they added, can transform a prompt such as “a comical R&B slow jam about a sock finding their match” into a “high quality, catchy track” within seconds.
The company says Lyria 3 improves on earlier models in three principal ways. It can generate lyrics automatically, giving users the option to create songs without writing their own words. It allows more granular control over stylistic elements such as genre, vocals and tempo. And it produces what Google describes as more realistic and musically complex compositions.
The resulting tracks are brief by design. At 30 seconds, they resemble social media soundtracks more than conventional songs. They are packaged with custom cover art generated by another model and can be downloaded or shared through a link. The company is explicit about its intent. “The goal of these tracks isn’t to create a musical masterpiece, but rather to give you a fun, unique way to express yourself,” the product managers wrote.
The feature also integrates with YouTube’s Dream Track, a tool for Shorts creators that uses Lyria 3 to generate custom soundtracks. By embedding generative music directly into short form video production, Google is strengthening ties between its A.I. models and its social video ecosystem. The move positions synthetic audio not as a novelty but as a built in layer of content creation.
At the same time, the expansion into music revives persistent concerns about authorship, copyright and authenticity. Google says all tracks generated in the Gemini app are embedded with SynthID, an imperceptible watermark designed to identify content created by its A.I. systems. The company has also extended verification capabilities within the app, allowing users to upload audio and ask whether it was generated using Google’s tools.
Google emphasizes that Lyria 3 is intended for original expression rather than imitation. If a user names a specific artist in a prompt, Gemini will treat it as broad stylistic inspiration rather than an instruction to replicate that artist’s voice or catalog. The company says it has implemented filters to check outputs against existing content and that it has been “very mindful of copyright and partner agreements” in training the model. It acknowledges that safeguards may not be perfect and encourages users to report content that may infringe on rights.
The rollout is limited to users 18 and older and is currently available in several major languages, including English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese. Subscribers to Google AI’s paid tiers will have higher usage limits, reinforcing a subscription model that ties expanded creative capacity to premium access.