A Scribe's Story
- By Jeffrey R. Wakefield
When Harvard professor Lani Guinier speaks Thursday night at Ira Allen Chapel, her topic -- "Diversity, Opportunity, and the Shifting Meritocracy in Higher Education" -– is bound to raise a host of provocative questions in the minds of audience members.
It’s a good bet she’ll also spur one she’s not intending: what magic, exactly, is responsible for the transcript of Guinier’s words that flows seamlessly in real time on the large screen next to the stage?
The magical outpouring, which accompanies events at UVM from convocation to commencement, with up to a dozen lectures in between, is the work of former court stenographer Norma Miller, stationed behind her steno keyboard on a raised platform next to the UVM Video team.
Miller is hired by UVM’s ACCESS office to make public lectures accessible to hard-of-hearing and the “late deafened” students and other UVM community members, who have never learned American Sign Language, and to non-native English speakers. She also attends class with some of these students, especially those taking linguistically challenging medical or graduate courses. Students sit next to her during class and read a transcript of the professor’s lecture on her laptop.
“She helps level the playing field, so students have the ability to receive information in an equitable fashion,” says ACCESS director Laurel Cameron. “We’re very fortunate to have Norma. She could live somewhere else and earn two or three times what she’s making here.”
Reduced to analogy
As the words pour onto the screen during a lecture like Guinier’s, it’s hard not to wonder how Miller does what she does. Don’t expect a simple explanation.
Stenography training programs, typically two years in length with a full year devoted to theory, often have drop-out rates of over 95 percent. Miller is reduced to analogy to explain what she’s doing: “It’s more like stroking chords on a piano than typing,” she says.
Stenographers reach warp speed -- Miller’s steady state is 240 words per minute, with 95 to 98 percent accuracy -- by using combinations of keys to create syllables, words or even groups of words. While there is a basic system at work, called machine shorthand, each stenographer creates his or her own unique keystroke variations.
In the old days, court stenographers would take their unintelligible steno transcripts and type them up into English. Today, computers handle the translation -- but not before the stenographer loads tens of thousands of common English words into the translation program, linking them to their unique steno language, a job Miller started with a marathon five-day session 20 years ago when she first began using the program. She continues the project before every lecture by adding new words the speaker -- whom she researches extensively -- is likely to use.
Serious steno game
Miller may look unassuming on her perch, pecking quietly away, but make no mistake: she has serious steno game.
Several years ago, on a whim, she decided to apply to the BBC, which was recruiting aggressively in the U.S. for live captioners for news, public affairs, and sports programming. One of only two American stenographers hired, she was captioning Wimbledon within six weeks of landing in London.
She loved the city, the people, and the work and stayed for 15 months -- with the network paying for frequent visits by her lawyer husband and daughter, who stayed behind in St. Albans.
“By the end, I was the person,” she says. When she announced she was returning home to reunite with her family, the BBC was desperate to keep her, offering to pay for her daughter to attend the prestigious American School in London.
Given the pressure she was accustomed to at the BBC, it’s not surprising that when UVM’s ACCESS office asked Miller if she would do real time transcriptions of public lectures -- where her words (and mistakes) would appear before a live audience, a situation few stenographers have the talent or stomach for -- she accepted without hesitation.
Her first UVM job, transcribing a highly publicized lecture by writer and Holocaust survivor Elie Wiesel, sent her scurrying to YouTube to familiarize herself with Wiesel’s accent and sotto voce delivery, and to a Yiddish/English dictionary to internationalize her computer program’s vocabulary. The job went seamlessly, and Miller has been a UVM regular ever since.
Miller’s work isn’t entirely error free, and some of her bloopers are priceless, like her transcription of the word “malnutrition” as “mounds of nutrition” during Cornel West’s recent talk, or the time she rendered the name “Chris Mahoney” in a speech by former Vermont governor Madeleine Kunin as its phonetic cousin, “charisma honey.”
Miller has as much fun as the audience does with her rare mistakes, especially when they take on a blue note, as when a slight keystroke error translated “no win game” in the same Kunin speech as “no win orgasm.”
Miller is confident enough in her abilities to accept hazardous duty steno assignments for comedians, a type of speaker most of her colleagues want nothing to do with.
One particularly fast-talking, stream-of-consciousness performer, the rising star Bo Burnam, was so hard to follow on YouTube that the ASL translators hired to sign for his show bowed out, leaving Miller as the sole translator for the deaf and hard-of-hearing communities.
Miller sent Burnham a note on his Facebook page letting him know she would be transcribing his performance. When he arrived on campus, he sought her out, promising to work her “mindfully” into his performance.
Mindfully or not, Miller found herself riding the verbal equivalent of a mechanical bull. When shocking language, jet speed deliveries, and words like “eucarist” and “antidisestablishmentarianism” didn’t throw her, Burnham had to admit he was impressed.
“Good effort,” Burnam said, standing next to the screen. “Look at her go, look at her go. This is so post-modern.”
Will the day come when voice recognition software renders the talents of Miller and her colleagues obsolete? That’s still a ways off, Miller says, since the software performs only at 90 percent accuracy for one voice the program is trained laboriously to recognize.
“UVM is such a shining star,” Miller says about the university’s commitment to providing access to students with disabilities through services like one she provides, which is rarely offered at colleges outside enlightened centers like Boston, New York, and Washington. Last year, Dartmouth College contacted Miller to transcribe commencement on their Jumbotrons. “They found out about that because of UVM. It’s a big deal. I think UVM should be very proud.”