South Korea-headquartered leisure big HYBE is leaning an increasing number of into synthetic intelligence in music.
The corporate has simply unveiled a digital pop group known as SYNDI8, whose voices have been developed utilizing AI expertise.
The expertise used for the group’s voices was developed by AI audio firm Supertone, which is majority-owned by HYBE, after it struck a $32 million deal for the tech in 2023.
Based in Seoul in 2020, Supertone claims that its software program can create “hyper-realistic and expressive voices that [are not] distinguishable from actual people.”
This expertise aligns with feedback made by HYBE Chairman Bang Si-Hyuk in a 2023 interview with Billboard, wherein he stated: “I’ve lengthy doubted that the entities that create and produce music will stay human.
“I don’t know the way lengthy human artists may be the one ones to fulfill human wants and human tastes. And that’s changing into a key issue for my operation and a technique for HYBE.”
The brand new digital pop group, SYNDI8, pronounced “Syndi-eight,” is a four-member woman group, comprising digital members Canary, Nest, Goyo, and Raven.
HYBE’s Supertone says they exist throughout the digital world of ‘Nansy Land’.
NANSY can also be the title of Supertone’s basis mannequin – which stands for Neural Evaluation & Synthesis.
Supertone is led by AI skilled Kyogu Lee (Supertone President). In an unique interview with MBW final month, Lee defined that the corporate’s NANSY mannequin “serves because the spine of Supertone’s speech synthesis applied sciences”.
“NANSY has the particular potential to divide and re-assemble voice parts – timbre, linguistics, pitch, and loudness – individually and independently, producing natural-sounding voices with unparalleled realism,” Lee stated.
You possibly can learn the analysis paper for NANSY, right here.
In its announcement about SYNDI8, Supertone stated it goals to “discover the functions of [its] expertise within the music content material enterprise and to domesticate [its] personal voice mental property (IP) to exhibit the capabilities of [its] creations”.
HYBE beforehand showcased the probabilities of what it might do with Supertone’s expertise in Could final yr when it launched a single known as Masquerade from an artist known as MIDNATT – described as HYBE’s first AI-powered artist.
MIDNATT, the digital alter-ego of HYBE-signed artist Lee Hyun, used a prototype model of Supertone’s real-time voice-changing instrument Supertone Shift to “seamlessly” transition between his personal vocals and the vocals of a feminine singer on the observe Masquerade.
That very same observe was touted by HYBE because the “first-ever multilingual observe produced in Korean, English, Japanese, Chinese language, Spanish and Vietnamese”.
Most not too long ago, HYBE’s new digital group has launched a three-track album known as MVP, which options lead single ATOTA (brief for Any To One To Any), My Fantasy, and True Coloration, with the latter observe adopting a contemporary R&B fashion.
You possibly can hear the lead single, ATOTA, beneath:
“SYNDI8 is a digital artist group born from Supertone’s voice synthesis applied sciences, impressed by the corporate’s progress story. Embodying the fantasy universe, the group plans to have interaction with followers throughout varied platforms.”
Deja and Han, SYNDI8
An animated story titled Magic Voice Prism: The Journey Begins was additionally launched alongside the album.
“SYNDI8 is a digital artist group born from Supertone’s voice synthesis applied sciences, impressed by the corporate’s progress story. Embodying the fantasy universe, the group plans to have interaction with followers throughout varied platforms,” stated SYNDI8 co-producers Deja and Han.
HYBE additionally demonstrated the capabilities of its proprietary AI on its Q1 investor name, when HYBE’s CEO Jiwon Park consented to having his personal voice cloned utilizing Supertone’s ‘voice synthesis expertise’.
Final month, MBW requested Supertone President Kyogu Lee if the corporate’s tech shall be used to recreate the voices of superstars like BTS.
He informed us: “Whereas Supertone is theoretically able to creating an infinite variety of new and unique voices, in addition to recreating present voices, we’re dedicated to prioritizing the rights of all artists and creators, together with these beneath HYBE.”
Supertone’s work has additionally been featured in Disney+ and Netflix collection.
The corporate says it’s increasing its attain to help creators with instruments like its AI noise discount plug-in Supertone Clear and voice changer Supertone Shift, which launched in April.
SYNDI8’s debut arrives amid the ongoing debate about using AI to create and recreate voices within the music business.
Music Enterprise Worldwide
