YouTube is exploring the future of music remixing with the launch of a new AI-driven feature, allowing creators to restyle licensed songs for their Shorts. The feature, named ‘Dream Track,’ enables creators to customize a song’s mood, genre, and tempo, offering a new level of creative freedom.
Currently in a limited trial phase, ‘Dream Track’ is being tested by a small group of invited creators. To use the feature, creators select an eligible song, provide a prompt to guide the AI, and the system generates a unique 30-second remix. The remixed track can then be used in YouTube Shorts, a format that has gained significant traction in recent years.
The remixed soundtracks generated through ‘Dream Track’ are clearly labeled with attribution to the original song, and include a note that the track has been altered using AI. YouTube has emphasized that only songs from artists who have consented to have their voices used by AI will be included in the trial.
Some of the artists participating in the experiment include Charlie Puth, Charli XCX, Demi Lovato, and John Legend. However, it remains unclear which specific songs will be available for remixing and which music labels YouTube has partnered with to facilitate the project.
This new feature is part of YouTube’s broader strategy to bring more creative tools to the platform, particularly for creators producing Shorts. AI-powered customization tools offer users greater flexibility to engage with music and generate personalized soundtracks.
Reports from June revealed that YouTube had been in discussions with major music labels to gain permission to use their songs for AI training purposes. The ‘Dream Track’ initiative could serve as the starting point for a more widespread rollout of AI music customization tools, further transforming how creators interact with music on the platform.
While still in its experimental phase, ‘Dream Track’ provides a glimpse into the potential future of AI-assisted music creation, where creators can seamlessly blend original tracks with innovative twists, potentially leading to an entirely new genre of user-generated content.