Click to view the complete transcript of the episode

Ways of Knowing

 

The World According to Sound

 

Season 2, Episode 6

 

Sound Studies

 

[instrumental music plays]

 

Siri voice: Here’s a response from Wikipedia.org

 

Chris Hoff: This is the voice of Siri, Apple’s digital assistant.

 

Siri: A virtual assistant (VA) is a software agent that can execute various tasks or services for a user based on user input such as commands or inquiries, including spoken ones.

 

Hoff: It can provide directions, play tunes, make calls, set timers, send messages, and address any questions you typically use Google for.

 

Sam Harnett: Siri, what’s Steph Curry’s free throw percentage?

 

Siri: Stephen Curry boasts a free throw percentage of 93.3 this season in the NBA.

 

CH: When Siri debuted in 2011, the sole American English voice available was that of a middle-aged white woman.

 

Siri: I didn’t catch that, could you repeat it?

 

CH: Fourteen years later, a significant number of virtual assistants still mimic the voices of white women.

 

[montage of female voice assistants speaking]

 

Golden Owens: Everyone has a unique voice. You can glean much about an individual from their sound, but you can also understand a lot about technology based on its auditory characteristics.

 

CH: Golden Marie Owens, assistant professor of cinema and media studies at the University of Washington.

 

GO: Why does she sound so mechanical, or why does she have the tone of a white woman, or why is she feminine at all? Each of these questions can lead to a broader exploration into topics like history, the evolution of sound and technology, and can provoke deeper discussions about race, identity, and the implications of intentional choices.

 

It’s quite fascinating that the default voice for all these virtual assistants, at least in the U.S., is that of a white female. That’s the standard unless changed. Why? Just contemplating that why can open up multiple avenues that may be unexpected.

 

CH: One path that Golden explored led to an investigation of servitude in the U.S. After all, these virtual assistants are specifically created to do tasks for us, to assist us. They are essentially digital servants. The historical context of servitude in the U.S. is extensive, with slavery playing a significant role.

 

GO: On the surface, it appears to be a complete transformation because we hear these white women’s voices. However, when we reflect on what historically contributed to these preferences for certain types of servants, there arose a notion of comfort, an idea of something that can be controlled and dominated. In many respects, that notion is tied to whiteness, yet also deeply intertwined with conceptions of Blackness. Therefore, it is important to consider these intelligent assistants as service-oriented entities while recognizing the origins of service itself.

 

[instrumental music plays]

 

CH: Golden has researched how interactions with the female-sounding virtual assistants mirror the manner in which people engaged with Black slaves. Her inquiry began after observing how Amazon marketed its digital assistant back in 2014.

 

GO: It started with watching the inaugural commercial for the Amazon Echo and realizing, “Oh, there’s a curious comparison here.” For context, it’s a commercial featuring a traditional white family, where a little girl describes all the capabilities of the Amazon Echo. It’s 2014, just launched, and at the end, she proclaims, “With all the things Echo can accomplish, it has truly become a part of the family.” Immediately, I noted, “That’s very specific phrasing.” Such language has often been used to refer to servants, particularly Black servants, referred to as “part of the family, just like one of us,” for instance.

 

This realization led me on a journey to explore how deeply Blackness is linked to our perceptions of these virtual assistants. Amazon’s design guidelines, for years, included directives like: be flexible, be relatable, refrain from excessive talking, avoid volume, respond as people wish to be addressed. Such precise instructions for developers felt akin to language used in master-servant dynamics. It seemed that there was an intersection of Blackness and technology being overlooked, as these assistants facilitate tasks in our homes and workplaces, performing duties we prefer to avoid, while echoing historical connections to servitude and slavery.

 

[instrumental music plays]

 

CH: Opting to have these virtual assistants sound like white women serves to conceal those historical connections. Even though the way you engage with these digital assistants mirrors interactions with past servants, they do not resonate with those historical figures. Instead, they adopt a fresh, seemingly disconnected persona from the legacy of servitude. A white female voice carries its own cultural implications — not due to its inherent traits, but because of the racial significance attached to it. In America, voices were racialized in a similar fashion as physical attributes like skin tone, hair composition, and facial features during slavery. Auditory characteristics were just as pivotal as visual ones in classifying individuals — a practice that continues today.

 

GO: The way in which individuals perceive race is linked to the historical context of voice racialization throughout time. In the U.S., this dates back to the Antebellum era when, as Jennifer Lynn Stoever discusses in her book “The Sonic Color Line,” enslavers could often no longer discern visible differences between themselves and their enslaved due to extensive violence and racial mixing. Consequently, the only alternative for identifying who was Black was through auditory means.

“““html

That’s when we began developing various definitions of what constituted a Black voice and what represented a white voice. The white voice was perceived as clear, serene, composed, high-pitched, yet somewhat low-energy in certain respects. Conversely, Black voices were viewed as quick, loud, abrasive, rugged, and more expressive than white voices.

 

CH: It’s impossible to design a voice for digital technologies that circumvents biases about speech patterns. There’s no such thing as a “neutral” voice. When developing a product, considerable focus is given to its functionality and appearance. However, equal care is devoted to how the product should sound.

 

GO: Sound is typically crafted to be something we don’t consciously engage with as much. Particularly from a media studies perspective, there’s significant emphasis on the visual elements, which is understandable. We have films, television, streaming services, and now VR. Nonetheless, audio and visual aspects often collaborate in specific ways. In many cases, one cannot fully comprehend the visual experience without also grasping the auditory context.

 

CH: Within our sight-dominated society, sound is frequently overlooked. Our ability to focus on auditory stimuli is far less honed than that for visual cues. Sound studies aims to highlight this imbalance and re-establish the significance of the auditory. Vision may be the predominant sense, yet there’s much to discover if we redirect our attention to our ears instead of our eyes.

 

CH: Here are five texts that will enhance your understanding of sound studies as a mode of perception.

 

“The Sonic Color Line: Race and the Cultural Politics of Listening,” by Jennifer Lynn Stoever

 

CH: Stoever examines the connection between race and sound in the United States. For her, the ideologies of white supremacy rely heavily on what we listen to—not merely what we perceive visually.

 

“How Do Voices Become Gendered,” by David Azul

 

CH: This article contests the notion that the acoustic characteristics of human voices are solely a result of biological factors.

 

“The Race of Sound: Listening, Timbre, and Vocality in African American Music,” by Nina Sun Eidsheim

 

CH: Eidsheim investigates artists Billie Holiday, Marian Anderson, and Jimmy Scott to illustrate how listeners perceive race through the vocal timbres present in their voices.

 

“Multivocality,” by Katherine Meizel

 

CH: Much like identity, vocality—how one sounds—is dynamic. Meizel explores singers throughout history who have transformed their identities by engaging in what she refers to as “multivocality.”

 

“The Possessive Investment in Whiteness: how white people profit from identity politics,” by George Lipsitz

 

CH: A seminal work discussing the dynamics that not only prompt white individuals to maintain the status quo but also to participate in institutionalized forms of racial discrimination, termed “whiteness” by Lipsitz.

 

CREDITS

 

Ways of Knowing is a project by The World According to Sound. This season focuses on the various interpretative and analytical approaches within the humanities. It was created in partnership with the University of Washington and its College of Arts & Sciences. All interviews with UW faculty took place on campus in Seattle. Music contributed by Ketsa, Human Gazpacho, Graffiti Mechanism, Serge Quadrado, Bio Unit, and our collaborators, Matmos.

 

The World According to Sound is produced by Chris Hoff and Sam Harnett.

 

END