Tucked away in a dim recording studio in Seoul, a groundbreaking experiment in the global music industry is underway. Hybe, the leading K-pop music label responsible for the worldwide sensation, BTS, is leveraging artificial intelligence (AI) to blend the voice of a South Korean singer with native speakers in five other languages.
Hybe’s innovative use of AI has allowed singer MidNatt’s track to be released in six languages simultaneously – Korean, English, Spanish, Chinese, Japanese, and Vietnamese. While K-pop artists have previously released songs in English and Japanese along with Korean, this simultaneous six-language release facilitated by AI technology is a first, according to Hybe, and could potentially be adopted by more prominent acts.
“We would first listen to the reaction, the voice of the fans, then decide what our next steps should be,” explains Chung Wooyong, the head of Hybe’s interactive media division.
For the recording of “Masquerade,” MidNatt, who is fluent in Korean and has a limited proficiency in English and Chinese, sang the song in each language. Native speakers then read out the lyrics, and the two were seamlessly integrated using Hybe’s proprietary AI music technology, resulting in a track that sounds natural and authentic in every language.
Hybe’s novel use of AI represents the increasing influence of technology in the music industry, which is already seeing AI-generated song mash-ups trending on social media and the Grammy Awards introducing rules for AI usage.
“We divided a piece of sound into different components – pronunciation, timbre, pitch and volume. We looked at pronunciation, which is associated with tongue movement, and used our imagination to see what kind of outcome we could make using our technology,” Chung explains.
The incorporation of AI in the production process has allowed MidNatt to explore a broader range of artistic expression. “I feel that the language barrier has been lifted, and it’s much easier for global fans to have an immersive experience with my music,” he said.
The AI technology utilized in MidNatt’s song is a product of the Neural Analysis and Synthesis (Nansy) framework developed by Supertone, which Hybe acquired for 45 billion won (US$36 million) in January.
Although the technology requires “weeks or months” to fully process the pronunciation corrections at present, Hitchhiker, the producer of “Masquerade,” envisions the process speeding up in the future, opening possibilities for other applications such as interpreting in video conferences.
This innovative use of AI technology in the music industry lowers the barrier to music creation and has the potential to democratize the industry, opening up new avenues for artists around the world.
READ MORE: