Reading:
Use the DOI links in the bibliography below to retrieve the readings using the Columbia Libraries e-resources. We will read only a small section of the Tompkins; I will scan and upload the relevant pages here.
Courtesy links: [Tompkins, pp. 18-50]
Response:
In a short (150-200 word hard max.) response to these three readings, briefly describe a theme or topic common to all readings.
Additionally
There are thousands of tracks using speech synthesis. Recently, Porter Robinson’s album, Worlds, makes heavy use of Vocaloid software. Two contemporary examples of the vocoder include Taylor Swift, “Delicate” (at the beginning) or Zedd and Maren Morris, “The Middle” (at 0:36). In both cases, the vocoder is used to provide backing vocals supporting the lead singer (so the tracks include both the unprocessed “dry” vocals as well as the “wet” vocoded part). You should also have a look at this short excerpt from the 1939 demo of the Voder.
All three of these articles explore the idea of the Vocoder’s ability to approach, and its proximity to human-ness—in what types of situations is it good that the Vocoder is distinctly not human, and where does it fail because of its inability to sound human. Tompkins discusses how “during World War II, artifice […] was a weapon in itself.” The abstract sound of the vocoder allowed for SIGRUV to have unbreakable code because of its randomness, but the vocoder failed in other ways because it sounded scary with the pitch jumping around unmanageably, and because it was not designed around a woman’s voice. Bell discusses how because the Vocaloid didn’t have a timbral signifier for gender, it pushed away from the idea of individuals and was popular as a creative tool in online communities. Mills discusses how new technology for disabled people can end up to be more exclusive and less assistive, because of the ways it fails to sound like human speech.
I found the difference in emphasis of each excerpt notable. Though all three selections expound upon the mechanics and history of the vocoder, they seemed to discuss it in different contexts. Tompkins seems mainly concerned with the history of the vocoder and vocal synthesis and follows the innovations of Homer Dudley to the vocoder’s use in the military and in the music industry of the 20th century. Mills similarly looks at the history of the vocoder, specifically its predecessor the Voder, which manipulated frequencies through mechanical and electrical means, and expounds on the research behind this innovative machine. Bell looks at the vocoder in a more modern context, particularly the Vocaloid software and its variations and delving into the details of concatenative synthesis. These texts all raise questions regarding vocal synthesis and its role in communication, art, and entertainment.
The question that sort of arises when all the sources are taken into consideration is – are vocal reproduction technologies perceived as “human” or not? Usually, the answer is no: when used as prostheses, for example, the mechanical larynx is perpetually unable to reproduce human qualities; vocoders are also used in modern pop music, intentionally, to subvert common tropes of “human” masculinity by allowing their users to transcend the human. In the case of the original Voder, however, and the current Hatsune Miku, these machines are not entirely all “machine” either. Tompkins refers to the Voder in endearingly human terms, detailing its struggle to replicate the human voice, and referring to it as AT&T’s “awkward star.” This may be a sort of intentional subversion, but it is made possible by the idea that we can perceive something with its own unique voice as, at least at some level, humanizable. Hatsune Miku also demonstrates the human-machine duality of speech reproduction; while the software’s position as a sort of “tool” of musicians is acknowledged by the musicians themselves, “her” position as a real character that struggles with her position as nonhuman is also a real component that is considered by her fans.
These three readings all describe ways in which humans have used voice technology to push the limits of the human voice for many different reasons. Tomkins describes how, in the electro hip hop genre of the 1980s, black musicians used the vocoder to render their voices otherworldly. This music, fit for outer space, acted as an escape from racial oppression on earth. Bell explains how vocaloid software, created from recorded samples of human speech. While vocaloid was originally created to mimic the human voice as closely as possible, some artists have used the software to create music with pitch ranges and tempos that are not possible for humans to produce, therefore using the software as a way to extend the limits of human expression. Mills discusses the use of artificial larynxes and vocoders by those who do not have the ability to speak on their own. These prostheses paradoxically both enable and prohibit “normal speech” due to their unique timbres that drastically differ from natural human speech.
It seems that all three readings are trying to determine whether the human voice retains its uniqueness, or supposed reflection of something “natural” even when used through distorted medium. Or, rather, where the humanity of musical production is located- is it in the physicality of the human voice, or through human methods of sonic/physical manipulation? In Tompkin’s example of the vocoder being used by African-American musicians, it seems that the agency to manipulate is most important.
Each reading explores the relationship of vocal synthesizers to humans and our bodies, however, every author connects and compares the technologies in a different way.
When exploring the vocoder, it was interesting to see how the parts of the machine were an ‘analogy’ for parts of the human body. In “Media and Prosthesis,” Mills quotes Homer Dudler’s “Synthesizing Speech” which explains that a vocoder’s battery represents a person’s lungs, the relaxation oscillator represents vocal cords, and so on.
In How to Wreck a Nice Beach, Tompkins explains that a vocoder could synthesize a black voice sonically, but it could not recreate the experience of being a black person in America which is something that can be hidden in a human voice.
Finally, in “The dB in the .db: Vocaloid Software as Posthuman Instrument,” Bell discusses how the Vocaloid software emulates speech, but the synthetic timbre produced by the software can fail to seem realistic because of the methods used to create the programmable voice, such as recording individual diphones, vowels, and triphones from a person.
The three readings cover the history and application of speech synthesis by looking at the technology – – the voder/vocoder emulates the core technology that made telephone widely used around the world, while the Vocaloid software provided a new kind of automated instrument which made amateurs able to produce music without organic vocal sound. It seems that throughout the history the emergence of new technology is the key which redefines people’s expectations and explores new possibilities, whether in commercial industries or music.