Google adds music-generation capabilities to the Gemini app
Google announced Wednesday that it is adding a music generation feature to the Gemini app. The company uses DeepMind’s Lyria 3 music generation model to enable the feature, which is still in beta.
To use the feature, you describe the song you want to create, and the app will generate a song along with the lyrics. For example, you can ask Gemini to create a “comedic R&B slow jam about a sock meeting its match,” and the app will generate a 30-second song along with album art created by Nano Banana.
Google said you can even upload a photo or video, and the AI-powered tool will create a song that matches the mood of the media file.
The company said Lyria 3 improves on previous generation models, creating more realistic and complex music tracks. Users can also change and control other elements such as style, vocals and tempo.
In addition to rolling out Lyria 3 to the Gemini app, Google is also making the model available to YouTube creators through the Dream Track feature on YouTube. a tool allowing creators to create AI-generated tracks. Until now, the option was only available to YouTube creators in the US. But with this release, Google is expanding the availability of Dream Track worldwide.
Google said you can’t outright imitate an artist, but if you add an artist’s name to your prompt, Gemini will create a song in a similar style or vibe. (It’s not clear whether generation will make it easier for others to decode a particular artist’s musical style.)
“Music generation with Lyria 3 is designed for original expression, not for imitating existing artists. If you promptly mention a specific artist, Gemini will consider this as broad creative inspiration and create a song that shares a similar style or mood. We also have filters to compare the output to existing content,” the company said in a statement blog post.
Google noted that all songs created with the Lyria 3 model will have a SynthID watermark to identify AI-generated content. The company said this is also the case adding capabilities to identify AI-generated music with SynthID within Gemini. Users can upload tracks and ask Gemini if they are AI-generated.
Music Generation is rolling out to all 18+ Gemini users worldwide with support for English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese.
AI-generated music has created mixed feelings among artists and listeners. On the one hand, companies like YouTube and Spotify are adopting AI and signing contracts with music labels to monetize AI-generated music. On the other hand, AI modeling and tooling companies are facing lawsuits from the music industry over copyrights on the training materials. Platforms like Deezer have published tools to flag AI-generated music to combat fraudulent streams of this type of music.
