Gemini’s New Voice: Now Compose Custom Soundtracks with Lyria 3

Google Gemini has officially found its voice with the integration of the Lyria 3 music model, allowing users to transform text prompts or photos into custom 30-second tracks. This breakthrough beta feature empowers anyone to generate high-quality audio, complete with AI-written lyrics and personalized cover art, turning the Gemini app into a pocket-sized recording studio.

Your Personal AI Composer: A Deep Dive into Lyria 3

Launched on February 18, 2026, the Gemini app has evolved beyond text and images into the world of custom music generation. Powered by Google DeepMind’s Lyria 3, this feature is specifically optimized for the Indian market with native support for Hindi, alongside English and several other languages. Users can now create everything from a “fun Afrobeat track” to a “comical R&B slow jam” by simply describing a mood, a memory, or an inside joke.

The innovation extends to visual inspiration: by uploading a photo or video such as a snapshot of a dog on a hike, Gemini uses the content to compose a track with lyrics that fit the atmosphere perfectly. Each 30-second creation is paired with unique cover art generated by Nano Banana and is embedded with SynthID. This invisible watermark ensures that AI-generated audio can be identified and verified, maintaining a commitment to responsible AI development in the music space.

Global Competition and the Sound Revolution

The global AI music industry is rapidly moving toward integrated, verified content creation. While platforms like Suno or Udio offer longer song formats, Google’s Lyria 3 stands out through its deep integration with YouTube Shorts via “Dream Track” and its focus on “original expression” rather than artist mimicry. By working in collaboration with the music community and using filters to respect copyright, Google is positioning Gemini as a responsible tool for creators worldwide.

Trivia: Lyria 3 wasn’t built in a vacuum; it was refined through the Music AI Sandbox. This experimental space allowed Google DeepMind to collaborate with professional musicians to ensure the AI-generated tracks were musically complex and respected the nuances of different genres.

UDHQ Take: At Unbox Daily HQ, we see Lyria 3 as the bridge that finally connects our digital memories to our emotions. In India, where music is central to every celebration, this tool gives every user the power to be a composer. Whether you’re an entrepreneur in Mumbai creating a jingle for a new product or a student in Delhi sending a personalized musical “thank you” to a teacher, the value here is instant, high-quality creativity.

The real magic is the accessibility it brings to the Indian creator economy. You don’t need a music degree or expensive instruments to express a feeling; you just need an idea. By removing the technical barriers to songwriting, Gemini allows your stories to be heard literally. At UDHQ, we believe the added layer of SynthIDverification is a masterstroke, providing the trust and transparency necessary for AI to thrive in the Indian digital landscape.

Source: Google