Every time I think it might be a good idea to enable the voice-to-text feature on my phone, technology proves it's a bad idea.
Sure, I dictate into my iPad letters that are handwritten, typed in all-caps or that the scanner app on my iPad decides are in an alien language, but I review those letters and make corrections once I've put them into our system. Siri might have thought I said "Parmesan evangelicals," but I didn't (insert cheesy joke here); I said "farmers and evangelicals." And "Judo-Christian ethics"? Are they doing martial arts in church now? I don't remember what "pathology ice cream" was supposed to be, but I despair for artificial intelligence.
As bad as my typing in texts to friends and family members may be (especially when accompanied by my phone's autocorrect with a weird sense of humor), if I used voice-to-text, they'd have no idea what I was talking about ... not that they do anyway, but still ...
From experience I can testify that Siri and Google are afflicted with the same thing, and both mishear me, usually with strange and/or hilarious results. Maybe they're toddlers, teenagers or hyperpartisans who only half-hear (at most) anything they don't want to hear. Or maybe they're just experiencing the human phenomenon of "mondegreens."
I've written of mondegreens before, but I never tire of talking about misheard lyrics and words. The mondegreen got its name in 1954 in an essay in Harper's Magazine by writer Sylvia Wright:
"When I was a child, my mother used to read aloud to me from Percy's Reliques, and one of my favorite poems began, as I remember:
Ye Highlands and ye Lowlands,
Oh, where hae ye been?
They hae slain the Earl Amurray,
And Lady Mondegreen."
But there was no Lady Mondegreen; after the Earl of Moray was slain, they "laid him on the green." As Wright wrote in the essay, the point of mondegreens "is that they are better than the original."
They're certainly more fun. And I'm all about fun with words.
In a 2014 story in The New Yorker, Maria Konnikova wrote: "Mondegreens occur when, somewhere between the sound and the meaning, communication breaks down. You hear the same acoustic information as everyone else, but your brain doesn't interpret it the same way."
Oronyms--strings of words that can logically be divided in multiple ways--are one of the causes, she wrote. "One version that [cognitive scientist and linguist Steven] Pinker describes goes like this: Eugene O'Neill won a Pullet Surprise. The string of phonetic sounds can be plausibly broken up in multiple ways--and if you're not familiar with the requisite proper noun, you may find yourself making an error. In similar fashion, 'Bohemian Rhapsody' becomes Bohemian Rap City. Children might wonder why Olive, the other reindeer, was so mean to Rudolph. And a foreigner might become confused as to why, in this country, we entrust weather reports to meaty urologists or why so many people are black-toast intolerant."
Or why Jimi Hendrix excused himself to "kiss this guy," and Creedence Clearwater Revival wanted us to know "there's a bathroom on the right." Then there's Crystal Gayle, who let us know that "doughnuts make my brown eyes blue."
Crystal might want to see a doctor.
Homophones don't do us any favors either, leading to use of "lead" when it should be "led," or "deep-seeded" instead of "deep-seated." These mistakes, called eggcorns (an existing phrase is replaced by a plausible sound-alike phrase), tend to be more annoying than funny.
Really, what's funnier? "For all intensive purposes," or "the girl with colitis goes by"?
That girl, by the way, might want to car-pool with Crystal to the doctor's office.
I do worry sometimes, being a word nerd, that too much use of voice-to-text without correction might lead to linguistic change that will make my inner grammar grouch rear up. We don't want that. She's crabby all the time.
How many times do I have to see "bias media" when it should be "biased media"? (Even though "bias" in this usage typically means whatever the speaker doesn't agree with, regardless of evidence to the contrary.) "Bias" is the noun, and "biased" is the adjective; last I checked, nouns don't modify nouns. What will happen when people begin believing the incorrect words are the ones they should be using?
Do we all need the apps on our phones, tablets and computers to be the equivalent of high-end dictation software? Are we made of money? Most of us use the voice-to-text tools we use precisely because they're free. We're cheapskates and proud of it.
I've tried every which way to dictate in a manner that will cause Google or Siri or whatever app I'm using to accurately translate what I say, but it seems near-perfect diction does no better than sleepy mumbling. I could try to find a free app that actually works, but so far, no dice.
Or I could just type.
Nah. Too tired. And Siri will so nicely recite the lyrics to "Bohemian Rhapsody" if I get her started. That she understands. "Farmers and evangelicals," not so much.
Assistant Editor Brenda Looper is editor of the Voices page. Read her blog at blooper0223.wordpress.com. Email her at firstname.lastname@example.org.
Editorial on 02/19/2020