Experienced writers and editors know that proofreading your own work is risky business, no matter how good you are at doing it for others. That’s because your brain knows what you want to say even if it isn’t on the page or the screen, and will “helpfully” fill in the blanks for you. You’ll skim right past typos, punctuation problems, and even missing words.
While there’s nothing better than having another human read your copy, solo professionals sometimes don’t have the time or budget for that luxury. So I’ve enlisted a digital assistant to provide the next best thing.
My first line of defense against my own mistakes is to have my computer read my draft back to me. Simply listening to your own work being read out loud by another voice — even a virtual one — will help you spot many mistakes you might miss if you were just re-reading.
There are a number of ways you can get your machine to perform text-to-speech tricks. My personal favorite is Tom Bender’s Tex-Edit Plus, an elegant text editing program for the Mac with an intuitive Sound menu built right in. I particularly like how Tex-Edit highlights the sentence it’s currently reading, making it easy to pinpoint the location of any mistakes you hear. The cost is also very reasonable — a $15 shareware fee that you can pay at any time without having to put up with ads or crippled features.
Choose a computer voice that sounds as close to natural human speech as possible. The best voices are based on recordings of real people. On the Mac, my favorites are “Alex” and “Vicki.” I’m also a fan of the high-quality voices produced by Cepstral, which are made for Mac OS, Windows, and Linux. Their “David” is typically my first choice for proofreading.
When choosing a voice, you’ll want to be aware of what language, region, or use it’s designed for. Most writers can get by with a general-use voice, but specialized voices are available for fields like education and healthcare.
You might also try to match your voice to the audience you’re writing for. “Alex” and “Vicki,” for example, are both intended to represent 35-year-old adults, while Cepstral’s “Robin” simulates a child. When I write for a European audience I’ll often proof the copy using a voice called “Serena,” one of the Mac’s system standards for UK English. Some voices like Cepstral’s “Dallas,” have a bit of personality that can be useful if it matches your target readers.
I sometimes skip text-to-speech proofing when I’m in a hurry, but I usually regret it. It takes very little time, costs next to nothing, and can save you a lot of revisions — not to mention embarrassment.