How HYBE Used Voice Synthesis Tech to Translate This Artist Into Six Languages

today10/20/2023 2

share close

A little more than two years ago, HYBE invested millions into Supertone, an AI voice synthesis startup. Their relationship would allow for a collaboration with an existent artist that created a new kind of artist entirely — here’s how it happened.

An AI Alliance

HYBE’s relationship with Supertone begins in February 2021, when HYBE reportedly invests $3.6 million into it. By January 2023, their partnership expands when HYBE acquires Supertone entirely for a reported price of over $30 million. Little was then known about HYBE’s plans to integrate Supertone technology into its music empire (the company’s stable of K-pop supergroups includes BTS, Tomorrow x Together and ENHYPEN) other than that HYBE’s founder and chairman, Bang Si-Hyuk, told Billboard in his recent cover story that one of HYBE’s first integrations was referred to internally as “Project L” and scheduled for May 2023.

The Digital Debut

Right on schedule, HYBE starts to tease an upcoming single from MIDNATT, a new alter-ego of popular Korean singer Lee Hyun. Two weeks later, on May 15, he releases his first single, “Masquerade,” using Supertone technology to help him translate it into six languages: English, Korean, Spanish, Japanese, Chinese and Vietnamese. Through Supertone, the singer improves his intonation and pronunciation of the languages to sound more like a native speaker. Shortly after its release, MIDNATT says that he was inspired by “talking about language barriers” with his team: “When I would listen to music in other languages, I couldn’t immerse into the music as well as in my native language, and we were talking about how we could overcome [that].”

How It Works

Voice synthesis is best understood as a subset of generative AI that lets users manipulate their voice while talking or singing, allowing them to assume the timbre and tone of a particular celebrity, character or loved one. But some companies, like Supertone, also enable users to make other edits, like altering the language, age or gender of speech.

The Results

It’s a controversial use of AI, raising ethical questions about assuming someone else’s voice and making one’s own unrecognizable, but MIDNATT’s employment of voice-synthesis technology was largely well received. “After I experienced it myself, I think it really depends on how you utilize it,” he says of using the technology. “The sense of responsibility is what matters the most. So as far as it is used in the music, I think it is a great opportunity for me to make [my song] more accessible and more immersive to the fans worldwide.”

This story will appear in the Oct. 21, 2023, issue of Billboard.

Written by:

Rate it

Previous post


Fyre Fest Fraudster Billy McFarland Sued By Prison Friend and Spurned Investor

Billy McFarland, the creator of the infamous Fyre Festival who served nearly four years in prison for fraud and lying to the FBI, is facing a new civil lawsuit claiming he ripped off an investor who gave him $740,000 for his new PYRT venture. In a summons filed in New York Supreme Court on Tuesday (Oct. 17), an attorney for 54-year-old Jonathan Taylor of New York — who met McFarland while both were serving prison sentences at Elkton Federal Correctional […]

today10/20/2023 3

Similar posts


Lana Del Rey & Quavo Preview New Song ‘Tough’: Listen

Is Lana Del Rey and Quavo‘s “Tough” almost here? The pair unexpectedly teased the new track Wednesday night (June 19), leaving fans of both artists shaking in the comments on Instagram. Quavo unveiled a clip of “Tough” that will be, well, tough to get out of your head once you’ve […]

today06/20/2024 1

Post comments (0)

Leave a reply

Your email address will not be published. Required fields are marked *