For one thing, I do not like my voice. Never have. When I listen back to recordings of it, I wince and wriggle at its sound. They call this phenomenon “voice confrontation.” Another contributor to my unease was that I’m not what you would call an audiobook person. I say this with absolutely zero shade to all of those audiobook people, who rightly and vociferously love and defend their medium against the reflexive haters, purists who dismiss audiobooks as diet reading. I would never!
It’s more that, as a music critic here at The Washington Post, I spend most of my waking and working hours in a state of prolonged and attentive listening. If anything, books offer some liberation from this listening, allowing my full attention to fall upon the page and catch the quiet current of a good paragraph, the silent trickle of words. My voice has no place in a scene this pleasant.
When I asked my book’s editor if I’d be the de facto narrator — exuding an enthusiasm that I now perceive as a performance strictly for my own benefit — she gently broke the “no.” Market research and reader feedback pointed clearly to a general preference among audiobook consumers for professional narrators, which I was not. I’m not sure I’ve ever so fully unclenched.
Instead, she said, I’d be sent a selection of audio files — short audition clips of potential narrators reading snippets of my book. My relief was now replaced with a different worry. After all, this book is rather personal. One could say gut-spilly, at times. The idea of enlisting another voice to — what, represent? approximate? mimic? — my own was almost as nerve-racking as the thought of reading the thing myself.
The book is called “Swole: The Making of Men and the Meaning of Muscle,” and it combines a memoir charting my own “twig to big” journey in the gym with a cultural history of the enduring desire among men to become beefcakes. But even the historical stuff is personal — plainly an attempt to pare down my efforts in the gym to a set of essential questions: Why do I do this? What about muscle is desirable and why? What do muscles mean?
But as I wrote the book, it also became very much a broader story about men, the many ways we contort ourselves to fit comfortably in the unforgiving form that is masculinity, and the role our bodies play in constructing and presenting our idea of what manhood looks and sounds like. But what does it sound like to question what manhood sounds like?
Choosing a voice for your story is sort of like choosing a typeface for your text. Imagine a novel set in Comic Sans (like some passive-aggressive note posted over the office microwave), and you have a reasonable analogue for what the wrong voice can do to a book. The comparison is even more apt when you consider the budding industry of AI-powered auto-narration services — which represent particularly tempting opportunities for self-published authors in search of low-cost solutions, publishers of dense technical texts or those seeking quick foreign-language translations.
Google’s current offering touts “best-in-class voice quality” from robo-narrators like Mary (a matronly librarian type perfect for bedtime stories or muffin recipes), Marcus (a bass-y papa bear with long tales to tell) and Michelle (a graduate student in an MFA program with an NPR internship and lots of checked facts).
Some of these artificial voices might be capable of a moment or two of deception, but then Mary’s cadence falters, or Marcus gives a word the wrong emphasis. To even the greatest literature, these well-meaning, well-read bots impart all the depth of a customer service menu.
My selection of real-life narrators was positively old-fashioned in its array of utterly human subtleties and suggestions.
There was A, whose cadence was a touch on the stilted side and whose breath seemed to surround each word like a shawl. Every sentence sounded like he was describing rich, thick, delicious milk chocolate. There was B — a clean, crisp, very district-manager-sounding guy who could easily sell an extended warranty on a new Ford Escape but maybe not the story of my first time kissing another man. And there was C, a gruff, world-weary ambassador for general Gen X malaise — all stubble and swagger and flannel. He didn’t seem to get any of my jokes.
And just as I was losing hope, there was D — or Mark, as I now know him. His voice existed in a familiar in-between: neither young nor old, neither masc nor femme, educated but uncertain. Anyone who has ever heard me talk — or read my writing — knows (and endures) my stubborn penchant for asides and parentheticals, like those ones there. Mark nailed them all: every cheeky pun or play, every buried gay nuance, every little glance to the camera. In just a snippet of my story, he sounded like he knew the whole thing.
When I was offered input on the book’s cover, my one plea was that it convey masculinity while also somehow undermining it. The designer rose to this challenge with a bold, beefcake-forward presentation (centered on 16th-century etchings of the Farnese Hercules) and a disarmingly soft color palette of buttery yellow, sea green and sumptuous magenta. Mark’s voice sort of did the same thing to the material: It was intellectual, musical, knowing and, in a book about men, subversive. He occupies the box of manhood while making clear a discomfort with the accommodations.
After the recording, Mark — full name Mark Sanderlin, an audiobook narrator based in New York — told me he was under no delusion that he was a shoo-in for “Swole.” The youthful tilt and gay lilt to his voice tend to informally disqualify him for nonfiction titles, for which authors often want a deep, authoritative tone.
Or, as he put it, “They want to hear it read as they wish they were.”
As a child in Oregon, Sanderlin did theater roles and background parts in films. He went to college in New York, where he studied music business with a concentration in voice. Working as a producer and a keyboardist, he accumulated a bunch of audio equipment; when the pandemic hit, a vocal booth he built in his living room became his new office as he leaned into narrating. It was the perfect career for a theater kid who grew up wanting to play all the characters in “Gypsy.”
Sanderlin observes the rise of AI narrators, and their potential to gobble up speaking roles for humans, with a bit of bemusement. Sure, they can read text, but they can’t feel it: Bots can’t register irony or sarcasm. They don’t get cultural references from my childhood. They can’t suddenly switch into Arnold Schwarzenegger’s accent and modulate it for contextual effect. They have no idea what to do with my scare quotes.
Instead, you get a uniform cadence and a distracting distance between voice and sentiment — 10 hours of which can feel downright dehumanizing.
Sanderlin estimated that it took him close to 20 hours to record my book — roughly two studio hours for every hour of finished audio.
“It’s tedious, physically demanding work that requires sitting in a hot padded box without moving for hours on end,” he said. “But in the midst of that, you also have to focus, stay very still, and have what’s coming out of your mouth sound natural and conversational.”
To Sanderlin’s ears, any capable audiobook narrator can narrate anything, but a deeper connection to the text tends to elevate the final product. And to my ears (and my inexpressible relief), Sanderlin’s does just that, taking my story from one body to another, just like a book might.
Somewhere in his voice lives an awareness of the spirit of the language that doesn’t need to be translated. And somewhere in my story, Mark found something that sounded like his own voice.
“I sometimes get in my head about whether I’m seen as just a ‘gay narrator,’” he said (scare quotes mine). “But I also feel a sense of pride that I get to be the voice of these books.” And then he takes the words right out of my mouth: “I am a gay man existing in the world. And I sound like this.”
Read More
