Futureverse, an artificial intelligence and metaverse technology and content startup, has launched JEN 1, which it touts as a “universal high-fidelity model for text-to-music generation.”
The launch came less than a month after Futureverse announced that it had raised USD $54 million in its series A funding round from backers including 10T Holdings and Ripple.
Futureverse was formed from the merger of 11 different companies late last year. The company last month said it will use the investment to scale the metaverse infrastructure and introduce AI to the metaverse.
Earlier in August, the startup launched a $50 million venture fund and studio called “Born Ready,” which the company says will focus on accelerating the development and adoption of emerging technology ventures that hold strategic collaboration potential with Futureverse or The Root Network, a public decentralized blockchain network optimized for metaverse apps and experiences.
Most recently, Futureverse unveiled JEN 1, a new text-to-music generation tool in a research paper authored by Futureverse’s Altered State Machine innovation team. The team includes Dr. Alex Wang, Patrick Li, Boyu Chen, Yao Yao, Allen Wang and Yikai Wang.
The paper delves into the complexity of music’s wide frequency range and the need for higher sampling rates to capture its nuances effectively.
Futureverse claims that its research paper and the method adopted by its researchers surpassed that of industry giants Google and Facebook.
In May, Google made public its MusicLM tool that can generate high-fidelity music from text prompts and humming. Google said the tool works by typing in a prompt like “soulful jazz for a dinner party”.
Separately, Meta recently launched AudioCraft, which consists of its MusicGen, AudioGen and EnCodec models. AudioGen was trained on public sound effects to generate audio from text prompts.
Futureverse says that its JEN 1 model has been evaluated by a panel of music industry A&Rs to assess its capabilities both computationally and through human assessments.
The company claims that early results indicate that JEN 1’s music output achieves a notably higher perceptual quality (85.7/100) compared to existing methods (83.8/100), Futureverse said, claiming that the model has already released preliminary audio demos.
The technology handles various music generation tasks with high fidelity, maintaining computational efficiency while operating at a 48kHz stereo audio quality, Futureverse said.
“Human sensitivity to musical dissonance demands high precision in music generation. 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,” said Shara Senderoff and Aaron McDonald, Co-Founders of Futureverse.
“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.”
Shara Senderoff and Aaron McDonald, Futureverse
Senderoff and McDonald have been investing in the web3 and blockchain space for over six years.
Prior to launching Futureverse, Senderoff co-founded music tech venture fund, Raised In Space, with Scooter Braun’s Ithaca Holdings and Zach Katz, former president of BMG in the US.
Senderoff has been featured on Billboard’s 40 Under 40, Women in Music, and Rolling Stone’s Future 25.
Meanwhile, Futureverse Co-Founder Aaron McDonald’s career spans 20 years in technology as an engineer, product developer, and as a businessman with portfolios worth over $1 billion.
Music Business Worldwide
