Musician and AI artist Holly Herndon has reinterpreted Dolly Parton’s iconic Jolene in a way we never expected.
Herndon has produced a video and single cover of the iconic tune sung by Holly+, the deep fake AI voice the artist modelled after her own. A programmer armed with a Ph.D. obtained at Stanford University’s Center for Computer Research in Music and Acoustics, the singer-scientist created a deep fake twin of herself last year. The Holly+ neural network is capable of processing any song to make it sound like she’s singing it, even while she’s not. This AI is the direct descendant of Spawn, the AI she and her partner Mat Dryhurst used to produce her 2019 album, Proto.
According to Pitchfork, “a modified score of ‘Jolene,’ comprised of new harmonies, was fed to Holly+ and then generated in Herndon’s voice. Accompanying the voice is Ryan Norris on guitar.”
What makes this work such an intriguing case in the world of AI art is that Hernon, now based in Berlin, is not using AI just to mindlessly reprocess something but to create something new that follows an artistic vision.
Herndon has expressed their forward thinking artistry for a minute and even announced the development of Holly+ back in July of 2021: “Vocal deepfakes are here to stay. A balance needs to be found between protecting artists, and encouraging people to experiment with a new and exciting technology. That is why we are running this experiment in communal voice ownership.”
Her musical version of DALL-E is a good example of how AI can be used to make something with true value and human sensibility, as opposed to just producing random stuff from a computer. Access to such software, perhaps, remains the obstacle to this growing trend but for now we look to Herdon for inspiration.
