You can now create 30-second music tracks on Google's Gemini AI

Lyria 3, Google DeepMind’s latest generative music model, was launched on Wednesday and is cited as the most advanced music generation model.

Users can easily access it by visiting the Gemini app, clicking the 'tool' bar, and choosing "Create music".

One can then describe an idea like “a comical R&B slow jam about a sock finding its match" or uploading a photo, and Gemini will translate it into a high-quality, catchy track within seconds.

Google encourages creatives to even push the envelope further by asking Gemini to take inspiration from something you upload.

Lyria 3 will also enhance the quality of each unique YouTube Shorts soundtrack. This makes it easy to quickly share with friends by downloading or simply clicking the share link.

"The goal of these tracks isn't to create a musical masterpiece, but rather to give you a fun, unique way to express yourself," Google said in a statement.

The launch has Lyria 3 available only in the United States and will be rolled out to YouTube creators in other countries.

Since its first launch in 2023, Lyria has gone through a series of recalibrations and training in collaboration with the music community to develop better technology.

Google says that after a series of engagements and experiments, like Music AI Sandbox, it has led to perfect training of Lyria 3. 

Google has also cautioned users to avoid prompting track generations using names of specific artists since Lyria 3 is designed for original expression, not for mimicking existing artists.

If your prompt mentions a specific artist, Gemini will take it as broad creative inspiration and create a track that shares a similar style or mood. 

There are filters in place to check outputs against existing content to prevent violations of others’ intellectual property and privacy rights.