Massive Ocean, claimed to be the “world’s first deaf Ok-pop group”, launched their new single utilizing voice cloning know-how from Supertone, HYBE’s AI audio subsidiary.
The monitor, known as Bucket Hat, premiered on the UN’s AI for Good World Summit in Geneva final week (July 9) earlier than releasing on streaming platforms, The Korea Herald reported.
Supertone’s Controllable Voice Conversion system was used to course of vocals from Massive Ocean members PJ and Jiseok, extracting their vocal timbre and mixing it with fluent English pronunciation from a information vocal, in response to the report. The know-how permits the artists to protect their distinctive sound whereas bettering readability for worldwide audiences, the report stated.
In a LinkedIn put up on Monday (July 14), Supertone stated: “We labored on enhancing their English pronunciation, preserving Massive Ocean’s distinctive vocal tone and elegance as they share their music with a worldwide viewers.”
“We labored on enhancing their English pronunciation, preserving Massive Ocean’s distinctive vocal tone and elegance as they share their music with a worldwide viewers.”
Supertone
AI for Good stated: “Jiseok, PJ, Chanyeon reside with various levels of deafness and their dream to turn into an idol appeared unpromising till AI was adopted.” The group was shaped April 2024 they usually have since produced and carried out music utilizing a mixture of AI instruments and tactile know-how reminiscent of vibration-based gadgets for choreography, The Korea Herald stated.
Their debut monitor Glow was acknowledged as the primary Ok-pop music to function AI-generated vocal assist, in response to the information outlet.
The group has already gained recognition being named by Billboard because the “Ok-pop Rookie of the Month” in September, whereas Forbes included them on its “Asia’s 30 Beneath 30” in leisure in Might.
Throughout a presentation on the UN AI summit, Haley Cha, CEO of Parastar Leisure, stated: “Our purpose was to permit Massive Ocean JJ to speak naturally with world followers, with out sacrificing their vocal character.”
“Supertone’s AI allowed us to separate tone and pronunciation, combining the strengths of each the artist and the information vocal,” Cha was quoted by The Korea Herald as saying.
Supertone CEO Lee Kyo-gu stated: “This collaboration reveals how Supertone’s AI know-how can assist new inventive challenges in Ok-pop.”
“Our purpose was to permit Massive Ocean JJ to speak naturally with world followers, with out sacrificing their vocal character. Supertone’s AI allowed us to separate tone and pronunciation, combining the strengths of each the artist and the information vocal.”
Haley Cha, Parastar Leisure
“We’ll proceed serving to artists push previous bodily and technical limitations of their artistic work,” Lee stated, in response to The Korea Herald.
Massive Ocean is ready to carry out in Minneapolis, Las Vegas, Kansas Metropolis, New York, Washington, Raleigh and Atlanta by early August.
Supertone’s voice cloning course of makes use of its NANSY (Neural Evaluation & Synthesis) basis mannequin, which the corporate describes as “a unified neural framework designed to carry out voice-related generative duties.”
“By means of its built-in construction, it maintains voice traits consistency throughout technology and permits for the management of 4 particular person components of voice, enabling the expression of all human voices on this planet.”
Supertone has already established partnerships with Disney, Netflix, ZTX and ZEPETO, in response to its web site.
HYBE acquired Supertone in 2022, with plans to combine its voice-creation instruments into music. Lower than a yr later, HYBE-signed artist MIDNATT launched a monitor known as Masquerade, billed because the first-ever monitor to function AI-generated vocals in various languages together with Korean, English, Japanese, Chinese language, Spanish, and Vietnamese.
In April 2024, the AI firm launched Supertone Shift, a instrument that allows artists to change and clone their voices in actual time. The instrument can be utilized, as an illustration, to show a single artist’s voice right into a refrain throughout a reside efficiency.
HYBE additionally debuted a digital pop group, SYNDI8 final yr, utilizing Supertone-generated vocals for all performances.
The Ok-pop large demonstrated Supertone’s capabilities throughout an earnings name final yr, revealing that then-CEO Jiwon Park’s investor presentation was really delivered by an AI voice instrument that replicated the chief’s speech patterns.
Music Enterprise Worldwide
