 |
| (Image credit: Shutterstock) |
To celebrate the impending release of their new artificial intelligence music generator with voices, ElevenLabs is showcasing a few standout tunes on X.
The first Suno song to reach one million plays and Udio's introduction of hyper-realistic voices this year have demonstrated that AI music is one of the fastest-growing genres of synthetic material.
ElevenLabs is better recognized for its remarkably precise voice cloning and lifelike artificial voices. However, the company has been gradually adding more AI noise capabilities, such as sound effects.
While the music is still in early preview and only available to ElevenLabs employees, samples that have been provided thus far indicate a quality that is superior to Udio in a variety of genres.
Discover the Sound of ElevenLabs Music AI
Based on user-suggested prompts, ElevenLabs staff members provided samples of jazz, pop, dubstep, and country music along with the primary X account for the synthetic sound platform.
It is a significant improvement over the 30 seconds of Udio or the minute of Suno if the three-minute duration shown in many of the clips is the norm. A single cue might produce radio-ready songs instead of requiring the often erratic use of track extensions.
There aren't many examples—roughly a dozen have been released thus far—but each one has natural-sounding vocals and lasts for at least two minutes, often even three.
"Every song in this thread was generated from a single text prompt with no edits," stated ElevenLabs in a tweet. This is quite remarkable. The first single lived up to its expectations, titled "It Started to Sing" and classified as a "pop pop-rock, country, top charts song."
Why is this such a big deal?
It will be almost impossible to make a living in commercial music within 3 years. The only ways will be touring, patreon, bespoke composition, but even these will be greatly reduced and are already hard. Music is not just a job category, it is a way of expressing our spirit,… https://t.co/qgf7O6Ybm2
— Beardyman (@beardyman) May 10, 2024
Reposting a dubstep song by ElevenLabs, English music producer Beardyman declared that it "will be almost impossible to make a living in commercial music within 3 years."
"We are not as special as we thought… art is a process… self-expression is a therapeutic activity… but none of that matters now…"
It's a tremendous accomplishment to be able to create a three-minute mixed pop music that is chart-friendly with only one cue. It may not be in the charts just yet, but not too long ago, a song recorded in a person's bedroom became the first to reach the top.
In addition, Beardyman questioned if, in the end, a clubgoer would really care whether artificial intelligence (AI) "made the bangers they're dancing to" as long as the LED "showing AI images" is still working. He also proposed the concept of a holographic DJ.
AI music is not going away. There will be certain copyright obstacles to overcome, especially with regard to the ownership of training data and the opt-out rights of musicians. It will still have access to a sizable collection even if its training is limited to licensed and non-copyright music.
Ultimately, it will be utilized in the same method as any other musical innovation: as a means for musicians to manipulate sound in ways that aren't now feasible (think synthesizers, drum machines, and sampling); and as an inventive approach for regular folks to produce enjoyable music to enjoy with companions.