Dream Track is the latest AI experiment on YouTube

YouTube said that it has begun testing Dream Track for Shorts, a suite of experimental technologies that allows users to create 30-second soundtracks using generative AI. Dream Track was launched last year and is being developed in collaboration with Google DeepMind.

The streaming giant teamed with nine musicians, including Charlie Puth, Demi Lovato, John Legend, Sia, Papoose, T-Pain, Troye Sivan, and Alec Benjamin, to compose soundtracks using AI-generated voices.

YouTube's first Dream Track experiment allowed US producers to create soundtracks using their voices. They might compose a prompt and select from a carousel of AI-generated voices to create an own music for their Shorts.

Lyria, a music creation model developed by Google DeepMind, powers Dream Track. YouTube announced that as part of its latest experiment, it will allow a restricted group of creators to create instrumental-only soundtracks lasting up to 30 seconds. The business has stated:

Now, creators in the new experiment group can enter English-only text prompts + select a music genre (pop, hip hop, country, etc.) to generate 30-second, non-looping instrumental soundtracks. We’re excited to continue to ideate more Dream Track features that we hope enable deeper engagement between music fans and artists on YouTube.

YouTube, like many other digital businesses, is integrating AI into its offerings. Earlier this month, it tested an AI-powered tool called Jump forward, which allows users to jump forward to a frequently viewed area of a video by double-tapping the screen.

This AI-powered feature organizes and summarizes comments on YouTube videos, currently available on the mobile app for large comment sections in English. YouTube now requires creators to properly label their AI-generated content.