The characteristics of the voices of species that perform acoustic communication can be explained to some extent by their relationships with the voices of other species that live in the same habitat. The "unique" acoustic characteristics of the voice facilitate species identification and improves communication efficiency. Therefore, in the evolutionary process, each species that inhabits together in a common environment may have been under evolutionary pressure that makes "different" voices from each other.
With this idea in mind, the artist used genetic algorithms to construct a sound generation algorithm that allows 16 species of artificial life to evolve their "voices" to be as different in frequency and other characteristics as possible from other species.The programming was inspired by D'Arcy Thomson's classic writings, Bernie Klaus's discussion of acoustic niches, and Eri Honda-Sumi’s study of the voice of the genus Teleogryllus. Evolutionary computation and acoustic synthesis are performed in real time, and are constantly streamed from the server room of Institute of Advanced Media Arts and Sciences (IAMAS). The genetic information of each individual is a set of parameters that defines the characteristics of the voice of that individual. The fitness (dominance) of an individual, is calculated from the size of the difference between the genetic information of each individual and the genetic information of individuals belonging to species other than the one to which the individual belongs. In addition to the spectrogram of the generated soundscape, the video shows the Japanese standard time at the time of generation, the number of generations of evolutionary computation, and the average fitness of each species.
All genes are initialized every Monday at midnight. A programmatic "mass extinction" (a depletion of diversity) will occur from Tuesday to Saturday at midnight. Evolutionary computation is not performed on Sunday, and on Saturday, the final evolutionary form would keep chirping all day long. The above is repeated every week. The initialized gene produces a cricket-like voice, which is the theme of this work. The artist came up with the idea for this work after listening to crickets on the banks of the Tama River in 2018.
There was Voice in the beginning. Human language is said to be a system based on arbitrary relationships, but in biophony, the acoustic difference in voice between species is itself a resource and a “meaning”. No matter how huge the gap is between voice and language, we are also animals. The ladder to origin does not ascend towards the heights of heaven instead it keeps extending ‘downwards’.
This work was produced as part of the activities of the Time-Based Media Project at the Institute of Advanced Media Arts and Sciences (IAMAS).
Artist and sound designer, born in Aichi, Japan in 1992. His work includes algorithmic generative music, sound installations, and sound design for dance and film.
His work focuses on themes such as recording, sound and listening, and the origins of voice and language in the environment.
In 2021, he presented "117", an experimental cinematic installation that explored the synchronization and dissonance between hearing and seeing in visual experience, and between time and real time in repetitive media. In 2020, the short film "Humongous!" directed by Aya Kawazoe, for which he wrote the music, was screened in the Short Film section of Cannes Film Festival Critics Week. He has also started a lifelong recording project, "Graft the Voices", which will be released in 2021. In 2018, his computer music series "Silence Trade" was selected for the Contemporary Computer Music Concert. In 2015, he held a solo exhibition of electroacoustic music entitled “For Talking Water”.
Author : Nobuhiko HAYASHI
Web design : Koki INAYOSHI
Special Thanks: Hiroshi YAMATO, Yoshitaka OISHI, Mitsuhiro HIBINO
[Masahiro MIWA, Shinjiro MAEDA, Shigeru MATSUI]