Yamaha unveils first piano AI system
Yamaha Corporation has released footage of the world’s first artificial intelligence (AI) piano system, in the company’s latest foray into the world of live music AI.
The piano system, which made its debut at the Ars Electronica festival in Linz, Austria, is capable of playing any piece of music in the style of late pianist Glenn Gould. Music hologram production company Eyellusion has also expressed interest in bringing Gould back to life, in the form of a hologram tour.
At the festival, the system performed solo and a duet with pianist Francesco Tristano, accompanied by a trio of Bruckner Orchestra Linz members.
The system consists of a player piano and the AI software, which applies deep-learning technology to play any piece in Gould’s style with the aid of sheet music data.
It also includes Yamaha’s original AI Music Ensemble technology, enabling the system to analyse the performances of human pianists and play alongside them.
“To bring artificial intelligence into connection with music should be the beginning of a discussion that searches to expand and improve our virtuoso actions”
“To bring artificial intelligence into connection with music should not end in a competition, but should be the beginning of a discussion that searches to improve us and to expand and improve our virtuoso actions,” comments Martin Honzik, senior director of Ars Electronica Festival, Prix and Exhibitions divisions.
Brian M Levine, executive director of the Glenn Gould Foundation, recommends the project be “taken into the music mainstream” due to the “keen interest”, “great deal of attention” and “spirited debate” it will generate.
The AI piano concert marks Yamaha’s latest foray into live music AI, following the reproduction of the voice of Japanese singer Hibari Misora through its Vocaloid:AI singing synthesis technology.
According to Yamaha’s senior general manager of research and development division, Koichi Morita, the aim of such AI projects is to expand “the boundaries of musical creativity”.
“By sharing some of our ongoing results with music enthusiasts at Ars Electronica,” says Morita, “I feel we have taken another step toward realising these new possibilities.”
Get more stories like this in your inbox by signing up for IQ Index, IQ’s free email digest of essential live music industry news.
Yamaha vocaloid recreates voice of late singer
Yamaha Corporation has reproduced the voice of legendary Japanese vocalist Hibari Misora, through its trademarked Vocaloid:AI singing synthesis technology.
The public debut of the technology took place on the television programme Bringing Hibari Misora Back with AI, broadcast in Japan on 29 September by the NHK (Japan Broadcasting Corporation).
The NHK-led project used Yamaha’s technology to present a live performance of a new song by Misora, to commemorate the 30th anniversary of her death. A high-definition, 3D video reproduced the singer’s likeness for the rendition.
Vocaloid:AI is an adaptation of Yamaha’s original Vocaloid, released in 2003. The new version uses artificial intelligence to improve the reproduction of tonal changes, taking recordings of artists’ songs and speech as machine learning data.
“This new evolution of singing synthesis technology has illuminated new possibilities in music”
“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,” comments Koichi Morita, general manager of the research and development division of Yamaha Corporation’s technology unit.
“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.”
Vocaloids, or singing synthesisers, are being used increasingly across the entertainment industry. Hatsune Miku, a Japanese vocaloid embodied by a hologram of a 16-year-old girl, has sold out venues across the United States and is embarking on a 2020 European tour, playing shows at the O2 Academy Brixton (5,000-cap.) in London, La Villette (6,000-cap.) in Paris, Berlin’s Verti Music Hall (4,350-cap.), the Sant Jordi Club (4,620-cap.) in Spain and Amsterdam’s 17,000-capacity Ziggo Dome.
Artificial intelligence has also been used to recreate living celebrities, with “digital twins” being produced for Chinese girl group SNH48 and celebrity television hosts.
Get more stories like this in your inbox by signing up for IQ Index, IQ’s free email digest of essential live music industry news.