Google Launches Lyria 3 on Gemini

Google has introduced Lyria 3, a new generative music model integrated into Gemini, allowing users to create custom 30-second songs using text prompts or images.

The rollout positions Lyria 3 as part of Google’s broader push to expand creative AI tools within its Gemini platform, enabling users to generate original music complete with lyrics and cover art in seconds.

AI-Generated Music Through Simple Prompts

Developed by Google, Lyria 3 functions as a digital music generator embedded directly in Gemini. Users can describe a scenario, mood or concept, and the system produces a 30-second high-quality track that matches the requested vibe.

For example, prompts such as “a nostalgic 90s alternative track about a rainy day in Manila” can yield a fully arranged composition with vocals and thematic artwork. Users can also select genre preferences—ranging from upbeat pop to acoustic or other supported styles—and adjust elements such as tempo, vocal tone and overall energy.

Image-to-Music Generation

Beyond text prompts, Lyria 3 also supports photo uploads as creative input. A sunset photo from El Nido or a candid city moment can serve as inspiration, with Gemini translating visual mood into melody.

After generating a track, users can refine it further using additional prompts to tweak specific details, offering granular control over the final output.

Expanding Creative AI Tools

Lyria 3 reflects the growing role of AI-assisted content creation across markets like the Philippines, where short-form video remixes, digital art and AI-enhanced storytelling have gained traction in recent years.

By embedding generative music capabilities within Gemini, Google is positioning the tool as both an experimental creative space and a practical solution for everyday users. Students may use it to score multimedia school projects, while content creators can generate custom background music without relying on stock libraries.

The launch also highlights how AI models are increasingly designed to function as collaborative tools rather than replacements for human creativity—allowing users to guide output through prompts and iterative adjustments.

Lyria 3 is now available through Gemini, accessible via gemini.google.com.