Years after a stroke, an American country music star has created music once again using artificial intelligence.
Randy Travis is sixty five years old. During the 1980s and 1990s, he was a well-known country music musician who had received multiple Grammy awards over his career. But he visited the hospital in 2013 complaining of a cardiac issue and subsequently experienced a stroke. Travis struggled to talk after the stroke. Up till now, he has not created music since his stroke.
Using artificial intelligence, or AI, Travis debuted a fresh song on May 3. It’s titled Where That Came From. The song was penned by Nashville, Tennessee songwriters John Scott Sherrill and Scotty Emerick.
Travis’ music is produced by Warner Music Nashville, which Cris Lacy co-presesses. Lacy stated she offered “What if we could take Randy’s voice and recreate it using AI?” to Travis’s wife Mary.
“We were all over that, so we were so excited,” Mary Travis said about the opportunity to hear her husband’s voice in a new song. “All I ever wanted since the day of the stroke was to hear that voice again.”
Lacy discussed creating an artificial intelligence system able to replicate Travis’s voice with London computer programmers. They generated two prototypes. One model drew 42 samples while the other used 12 vocal samples from Travis’s career. There was enough of material to work with; Storms of Life, the first major record of the artist, released in 1985.
Kyle Lehning is Travis’ long-time music producer. Where That Came From was a song Travis would have fit in terms of vocals. Travis was renowned for a more conventional style of country music while he was at the pinnacle of his career. It was unlike the pop-country tunes that dominated the late 1980s.
Lehning discussed how the song was created. Another performer recorded the song in what is known as a “demo,” beforehand. The AI system then examined the song and overlaid Travis’s voice.
About the experience of seeing the computer replicate Travis’ voice on the song, Lehning remarked, “I really wish somebody had been here with a camera.” “And it was stunning, to me, how good it was… right off the bat.”
Lehning claimed that of the computer’s produced sound lacked “authentic resonance to Randy’s performance.” Lehning got it right working with recording engineer Casey Wood, who knew Travis well.
The two musicians combined vocal model sound elements. They changed the speed of several of the sounds. Lehning stated Travis is “a laid-back singer,” hence some passages have to be slowed down. Travis’s voice needed to retain a “old soul quality”.
Lehning reported they made progress on the artificial intelligence variant. Furthermore, Mary Travis claimed that the final form of her husband’s music differs greatly from what results from an AI creation done without authorization by a computer programmer.
She referred to the effort’s “human element” the labor of the engineers and producers.
She also mentioned that when her spouse initially heard the music, he seemed serious. “I believe he experienced every feeling there is in those three minutes of simply hearing his voice once more,” Mary Travis said.