Skip to content

YouTube adds text-to-instrumental feature in Dream Track

  • by
  • 2 min read

YouTube has extended Dream Tracks functionality, allowing users to create custom instrumentals for YouTube Shorts using English text.

Initially available to a select experiment group and supporting English text prompts, Dream Track enables users to specify music genres and receive instant, tailor-made compositions by artists who choose to participate. This streamlines the workflow and eliminates the hassle of sourcing music from multiple sources.

Artists like Charlie Puth, Demi Lovato, Sia, and T-Pain are working with YouTube to generate AI music for Dream Track. Currently, the tools support English-only prompts, which can disappoint many users.

“The first Dream Track test feature we launched creates soundtracks using the AI-generated voices of artists who have chosen to participate,” YouTube announced. “Now, creators in the new experiment group can enter English-only text prompts + select a music genre (pop, hip hop, country, etc.) to generate instrumental soundtracks for their Shorts.”

YouTube is gradually adding more and more AI-powered functionalities to Dream Track.

On May 3, YouTube experimented with a functionality that allows creators to use artificial intelligence (AI) for content inspiration. This tool, featured on the Studio desktop, aims to help content creators brainstorm ideas, understand audience dynamics, and create faster videos.

The development of Dream Track is powered by Lyria, an AI algorithm developed by Google DeepMind, the company’s AI research unit. In addition to Dream Track, Lyria has also given rise to another tool, Music AI, which empowers artists participating in YouTube’s incubator program to create, remix, and modify tracks innovatively, expanding the creative possibilities within the YouTube Shorts platform.

Google has implemented Synth-ID technology to address concerns within the artistic community regarding fair compensation and recognition. This technology adds inaudible watermarks to music generated using Lyria, enabling Google to race and credit artists whose work has contributed to the training of AI algorithms.

In the News: Threat actors target Indian elections with data leaks, campaigns

Kumar Hemant

Kumar Hemant

Deputy Editor at Candid.Technology. Hemant writes at the intersection of tech and culture and has a keen interest in science, social issues and international relations. You can contact him here: