The Replaced Voice
From the early existence of human being, art was a medium for the representation of nature - or the “known”- , in order to imitate the characteristics that define it or transform them to create something that triggers imagination1. As soon as the build of electronic instruments started to develop, there was a temptation to associate the electronic sound with the human voice. Within a century, this process of imitating or transforming the voice, lead to voice's complete replacement by machines.
This vision started about 1919, when the Russian physicist Lev Sergeyevich Termen - also known as Leon Theremin - invented an instrument called Theremin. Regarding its different filters (switches on the side of the instrument) that controlled timbre, Termen claimed that it could produce a “realistic imitation of the human voice”2. The composer Ernst Toch, characterised it as “rich, tempting, promising and…enchanting for the artist”3
Theremin was described as “a new voice to interpret serious music” and “a first step to future horizon”4. “All these experiments revealed, in embryo, the virtually infinite tone colour resources that would awaken in future synthesisers”5.
Termen gave the invitation to technicians, scientists and musicians for more research and technological achievements regarding electronic musical instruments related to human voice, from synthesisers or “choir organ” to the Vocoder, Auto-tune and vocal samples.
The Vocoder
During the decades of 1960s’ and 1970s’, post-war Germany radical artists were in the research of new music and aesthetics that deviated from the pop cultures of United States or England. The door was found through the electronic music. “Electronic music was virgin. Neither tainted by the past nor an Anglo-American import”6.
In 1960s’, Stockhausen’s musique concréte was along with the group of Popol Vuh that composed chorals with a synthesiser - Mellotron, as they called it choir organ7 - “that sounded like voices”8 for Werner Herzog’s films. At the same time, Joachim Roedelius, after experiencing the inhumanity of the Third Reich, formed Cluster that pioneered ambient electronica and aimed for a “sonic utopia”, promising a way out of the surroundings9.
Then, Kraftwerk appeared. The post-war German experimental ensemble in Düsseldolf of the 1970s’, shared the same vision and provided the link between German avant-garde and popular music10. The two founding members, both classically trained musicians, Ralf Hütter and Florian Schneider-Elseben, had an instant goal to produce music that sounded as if it had been created by machines11.
Schneider said in a documentary: “The human body has a small electric current - you can see on an ECG that there is no separation between humans and technology, for us they belong together a s unity”12. “With electronic music, humour and romanticism”13 Kraftwerk brought The Robots (1978) into the scenery and sound of music. Voice had an integral part in this goal14.
The invention and use of the Vocoder (vocal synthesiser) originated back to 1937, when Homer Dudley developed the “Voder” (Voice Operation DEmonstratoR), one of the first devices capable of imitating human voice that could produce “high-quality machine-synthesised sentences”15. Continuing his research, in 1939 he developed the Vocoder (VOIce enCODER). It was originally designed as a cryptographic tool, to provide secure voice communications over telephone lines during the Second World War16. Later on, in the late 1960s’, the Canadian electronic music composer Bruce Haack built the first Vocoder specifically for musical use and Robert Moog, the inventor of the analog synthesiser, followed with his own version17 .
Voice (input) is analysed in periodic sound signals rich in harmonics. Briefly, the Vocoder divides these sound signals into narrow frequency bands, then analyses the level of each band and gives an instantaneous vocal signal representation which can accurately reproduce the spectral content of the voice. The original frequency bands are reconstructed for the final sound using filters and modulators18.
Kraftwerk commissioned a full custom Vocoder which was firstly used in their album Autobahn (1974). The robotic voices formed a whole new aesthetic. Later on, during the 1990s’, Schneider patented a new system with a method of synthesizing singing in real time, controlled by a MIDI and a phonetic keyboard. These machines were the medium for the creation of the albums “Man-Machine” and the “Computer World”.
While performing The Robots, the band members were in various ways replaced by robots that resembled them. Not only voice is reproduced by the Vocoder, but the performers themselves are replaced by machines. In 2000s’ their performances developed into fully robotic shows.
Kraftwerk’s robotic voices and performances opened a new gate in the world of art and raised many questions regarding the future, its music and art. “The shiny new bodies and mythic metamorphoses enacted in metallic performance art celebrate but also forewarn of the gradual disappearance of the human body. Intelligent and conscious machines in the future might take their origins from humans, but they may equally originate from other sentient machines.”19
Auto - Tune
The rapid progression from the phonograph and analogue tape machines to digital recording softwares, instruments and sampling was the “ultimate outrage" that took-over during the 1990s’ . The conversion from analogue to digital was almost complete with the available computer-based software20.
Auto-Tune is a software program that was originally created to correct the pitch of vocal tracks in 1977 by Harold Hildebrand. While in popular music it was mostly used for pitch correction - creating big debates regarding what makes the “good singer”-, composers found this vocal electronic instrument a "synthetic feel”21 in studio or in live performance.
Yannis Kyriakides used Auto-Tune in his work Lunch Music (2013), for three male voices, three percussionists and live electronics. Written for a live dance performance, the three singers used live-midi controlled auto-tune, singing in the microphone while playing on the keyboard the desired pitches.
Inspired by William Burroughs’s book Naked Lunch, the composer describes: “The concept of the music revolves around the idea of the polyphony of voices in Naked Lunch. Voice deconstructed into sound, text, utterance, and communication gesture. There is often the effect used of one voice speaking through another, in how the live electronics controlled by the percussionists is modulating source material. Just like in the narrative sense in Burroughs work, the voices in this piece are transmitting through interchanging persona, and in contrast to the corporeality of the subject matter are often presented through some form of disembodiment.”22
Auto-Tune’s use is a medium that encompasses words such as “deconstruction”, “sound”, “interchanging persona” and “disembodiment” of the voice. The polyphony is not created by human voices but with a digital instrument.
There is something fascinating in these unnatural voices or sounds that remind of voices.
Regarding my own research, two years ago, exploring the complete voice replacement by Auto-Tune, I composed Give them rest (2020). Layers of auto-tuned voices interact and catharsis comes in the end with a virtual orchestra and myself singing Pie Jesu by Gabriel Fauré.
All the voices use Auto-Tune, except for the one who sings Pie Jesu. The idea was to limit the pitches to A (mainly), E and G# in a way that different material in different octaves create a kind of distorted isocrates (bourdon) while a close up voice asks questions and other playful ones accompany it.
In that time I was listening to Laurie Anderson’s O Superman (1982) that influenced and inspired me a lot, regarding the robotic voices and the use of the Vocoder. Inspired also by a friend’s experience, who - under the influence of Ayahuasca - saw her soul (a playful one) and discussed with it. I wanted to create this somehow mechanical and unhuman vocal soundscape asking questions to the subconscious in both an innocent and an almost critical manner.
I find these instruments, as a composition tool, to be a source of fascination and attractiveness. They transcend human voice, human speech or singing, penetrating into an unknown area or era that gradually -with increasing technological development - becomes a norm in our everyday life. At the same time, their use creates questions. Will music need human voices after all? Will they dominate the natural voice in our “developed” civilisations?
Vocal Samples
The questions become more intense when vocal samples manage to produce so well the quality of human voice that the limits between real and virtual blur. A vocal sample is a previously recorded voice, that can be bought, downloaded and reused in workstations/softwares for music production. They can be manipulated regarding various filters, looping, layering, speed and duration or pitch, according to the composers’ choice23.
With samples, composers can manipulate whole virtual choirs and orchestras in their home studios. Different male or female voices with different timbres and characteristics, answer to any request for music creation. While there is some diversity of voices that can be accessed in sample packs, the industry tends to cater for the conventional ideas of voices connected to a particular genre.
In our co-produced album Silent Lands (2020) for voice and electronics, in the work Under the Volcano the composer Andreas Moustoukis used male vocal samples to create a polyphony in many layers. Giving the sense of Indigenous people’s voices, the work visions an ensemble singing in the deep ancient forests. Above the electronic sounds, mens’ low voices sing with various pitches, timbres, layers and durations.
Their quality confuses the listener. One most probably will understand that these are vocal samples but it doesn’t make any difference in the value of the composition. Instead, it gives a whole new aesthetic in the contemporary approach to the voice.
The Replaced voice becomes an element of a new digital or virtual world.
Gut Thoughts from Lunch Music (2013) by Yannis Kyriakides.
Using Auto-Tune, the performers are playing the pitches that create the rest of the pitches while they sings. Another typical example of the use of Auto-Tune in Lunch Music is "Boy" (second track in the album).
Clara Rockmar performing Rachmaninoff's
Song of Grusia (op.4 , no.4)
fragment from the documenary Clara Rockmore: The Greatest Theremin Virtuosa, produced by Robert Moog and Big Briar, in 1998.
Aguirre (1975) by Popol Vuh
with the use of "choir-organ" (synthesiser)
written for Wermer Herzog's movie Aguirre, der Zorn Gottes" (1972)
Kraftwerk performing The Robots(1978)
The video (edited for research purposes) presentes the development of their performance regarding the replacement of performer's physical presence by robots, from 1978 in a French TV performance (performers are present imitating robots) to 2009 during their tour (exclusive replacement by robots).