Moving a step closer towards augmented eternity, Yamaha Corporation, in association with NHK (Japan’s national Broadcasting organization) used Artificial Intelligence (AI) to recreate the performance of a renowned female vocalist who passed away 30 years ago.
They used Yamaha’s VOCALOID: AI, a singing synthesis technology to bring back Hibari Misora for a new song ‘Arekara’ in a program aired on September 29. It was the public debut and first real-world execution of VOCALOID: AI, Yamaha officials on Tuesday.
Misora had sung for NHK for decades and acted in over 160 Japanese movies. On the year of her recording debut itself, she sang a solo that sold a record 450,000 copies. Nearly a decade after her death she was still voted as one of the most influential singers in Japan. NHK paid tribute to Misora on her 30th death anniversary by using 4K 3D video to reproduce her likeness. Misora wore a cream-colored dress in the video, took to the stage and left the viewers amused at the originality of the video.
How It Works
The fete was achieved by feeding the original recordings of the artist’s songs and speeches to machines. The tone and expression within recordings of singing in any language can be picked up by VOCALOID: AI. AI can synthesize singing which includes the unique mannerisms and nuances. Yamaha’s accompaniment suppression technology weeds out the background music to clearly extract the vocals.
“We believe it was the Yamaha technologies and sensibilities cultivated over 130 years of developing and producing musical instruments and audio equipment which enabled us to successfully capture the essence of her singing,” said Koichi Morita, Senior General Manager of Research and Development Division, Technology Unit, Yamaha Corporation.
“Our cooperation in this project with this new evolution of singing synthesis technology has illuminated new possibilities in music by transcending the barriers of time to dazzle listeners with incredible singing,” he added.