The wait is over and the rumors were true: Generative music startup Uncharted Labs launched its own music generator today, called Udio. It transforms simple text prompts into professional-quality tracks, and is poised to give Suno V3—the current king of AI music—a run for its money.

Before its official release, a few songs generated by Udio were leaked by anonymous sources—and the quality was impressive, to say the least. With coherent lyrics, well-structured compositions, and a rhythm that could rival any human-composed track, some reviewers have already hailed Udio as the best song generator to date, beating Suno V3 at its own game.

The first teases appeared last weekend, with some accounts from the AI community sharing leaks of a hip-hop song and another vintage-themed song with a very capable female voice.

The company remained silent at the time, but Udio now appears to have proven the rumors true, sharing high-fidelity samples that match the leaks in quality.

"Udio builds AI tools to enable the next generation of music creators," the company states on its official website. "We believe AI has the potential to expand musical horizons and enable anyone to create extraordinary music."

The platform, which is now open for public testing, has already seen a massive influx of users eager to try out the new technology. However, this popularity has led to overloaded servers, resulting in a wait time of over 30 minutes to generate two songs. Just for comparison, it takes around 90 seconds to generate a song with Suno V3. This issue, however, could be normalized in the coming days as traffic stabilizes.

Decrypt tried to generate some samples to compare them against SunoV3, but the platform failed. It even throws a cute screen with a nice error-themed AI song when it collapses.

Udio's interface is user-friendly and intuitive. Users can create music by simply typing a description of what they'd like to create in the prompt box. For example, typing "a country song about Decrypt Media" will generate a track that fits that description. Like other models, Udio also allows users to add tags like "mellow" or "warm" to further customize tracks.

Udio is backed by a16z, Black Eyed Peas musician Will.i.am, Tay Keith, Steve Stoute’s UnitedMasters, Mike Krieger, and Kevin Wall, among others. The company says it’s committed to empowering artists, and has some legal safeguards in its terms of service to assure both better model training and ethical use of its technology.

For example, users allow Udio to train future models with their creations, and the terms even restrict people from streaming other users’ content on social media and music platforms like YouTube, Instagram, TikTok, SoundCloud, or Spotify. The platform, however, lets users commercialize their own creations.

Udio will remain free during its beta testing phase, with users able to create up to 1,200 songs per month —when it works, that is.

Edited by Andrew Hayward

Generally Intelligent Newsletter

A weekly AI journey narrated by Gen, a generative AI model.