On the heels of a $54 million funding round, artificial intelligence and metaverse company Futureverse has officially introduced its JEN-1 text-to-music AI tool, which it says is superior to both Google’s MusicLM and Meta’s MusicGen.
Auckland-based Futureverse reached out to Digital Music News with its formal JEN-1 announcement as well as a corresponding white paper about the text-to-music generator. Said white paper describes in detail the purported technical advantages of JEN-1 and indicates that its AI music is of a “perceptually higher quality (85.7/100) compared to the current best methods (83.8/100).”
For background, Futureverse in December merged “eight metaverse companies into one collaborative ecosystem,” and higher-ups promptly released a separate white paper about their long-term vision for the overarching operation.
Among the eight companies that combined within Futureverse were AI developer Altered State Machine and Web3 payments platform Centrapay. Now equipped with 300 or so full-time employees, according to execs, the entity counts as co-founders Shara Senderoff (who also co-founded Raised In Space with Scooter Braun), CEO Aaron McDonald, CTO Marco Brondani, and chief business officer Daniel Gillespie.
Returning to the newly detailed text-to-music generator, Futureverse’s north of 5,000-word white paper on the forthcoming technology reads in part: “More importantly, human judges confirm JEN-1 generates music highly aligned with text prompts in a melodically and harmonically pleasing fashion.”
These “human judges,” the company behind JEN-1 elaborated, are “respected music industry A&Rs to be announced soon.” Moreover, Futureverse has made publicly available a number of JEN-1 instrumental samples, including alongside the brief AI-created works the underlying text prompts and outputs from competing text-to-music services for the same prompts.
“Human sensitivity to musical dissonance demands high precision in music generation,” Senderoff and McDonald said in joint remarks. “We have been working deeply in this space for the last two years. We are incredibly proud to publish a first look into our team’s significant progress in the advancement of music AI that will benefit creators and progress in the music industry.”
While the music attributed to JEN-1 appears technically solid, it goes without saying that the tool (and all other available text-to-music artificial intelligence offerings) represents the tip of the AI-music iceberg.
As the unprecedented technology has rapidly evolved to this point in 2023, so too have the AI approaches of the Big Three record labels. With firmly worded public comments and more than a few takedown notices having failed to stem the tide of soundalike AI music – authorized AI music is likewise becoming increasingly prevalent – Universal Music in May partnered with relaxation app Endel to create tracks using “ethical AI.”
Moreover, the leading label and Warner Music are reportedly in talks with Google to license their artists’ voices for an AI music-generation tool. Additionally, Warner has teed up the release of artificial intelligence projects from “big names” for later this quarter.