- Google has introduced Lyria 3, a new AI music model now available in the Gemini app.
- Users can create 30-second songs by typing a prompt or uploading a photo or video.
- The tool generates lyrics automatically and allows control over style, tempo and vocals.
- All tracks include a hidden SynthID watermark to confirm they were made using Google AI.
Google has expanded the capabilities of its Gemini app with the launch of Lyria 3, a new artificial intelligence model focused on music creation.
The update allows users to generate custom songs within seconds by simply describing an idea. A playful prompt, such as a funny love song about a lost sock, can instantly turn into a polished R&B track. Users can also upload images or short videos, and the system will compose music that matches the mood of the content.
The goal, according to Google, is to give people a simple and fun way to express themselves through sound.
Lyria 3 builds on earlier versions of Google’s music tools by offering stronger audio quality and greater creative freedom.
Users do not need to write their own lyrics, as the system creates them automatically based on the prompt provided. At the same time, people can guide the output by choosing elements such as genre, tempo, and vocal style.
The result is a 30-second track complete with custom cover art. The artwork is generated automatically, making it easy to download the song or share it directly with friends.
The feature supports both text and media uploads. Someone feeling nostalgic can request a song about childhood memories in a chosen genre. Another user can upload pictures from a family outing or a pet’s adventure, and the app will turn those moments into a soundtrack.
While the tracks are short, they are designed to capture personal stories in a quick and creative way rather than produce full studio albums.
Beyond the Gemini app, Lyria 3 is also being integrated into YouTube’s Dream Track feature. The update will help creators produce improved background music or short lyrical pieces for their Shorts videos.
The rollout began in the United States and is gradually expanding to other countries, giving more creators access to AI-assisted soundtracks.
To ensure transparency, every song created with Lyria 3 contains SynthID, a hidden digital watermark that identifies content generated by Google AI.
Users can also upload an audio file into the Gemini app and ask whether it was made using Google’s AI tools. The system checks for the watermark and provides feedback.
Google says the technology was developed with careful attention to copyright and agreements within the music industry. The company emphasises that the tool is meant to inspire original creations rather than copy existing artists. If a user mentions a well-known musician, the AI will draw general inspiration from the style without imitating specific work.
Filters are also in place to prevent content that may violate intellectual property rights. Users are required to follow Google’s terms and policies when using the feature.
The music generation tool is available to users aged 18 and above in several languages, including English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese. Google AI Plus, Pro and Ultra subscribers receive higher usage limits.
With this latest update, Google is positioning the Gemini app as more than a text assistant. It is now a creative companion, offering users a simple way to turn ideas, images and memories into personalised soundtracks.
