Commentary"Linked Up" Column
from the December 13, 2006 edition

Why more folks may talk to their computer

The latest version of Dragon NaturallySpeaking makes dictating emails and term papers easier than ever.

It's become a science-fiction cliché, at least among Star Trek fans: Captain Picard (or Kirk, or Janeway, depending upon your favorite series) needs a complicated task accomplished. "Computer," Picard says, "give me all the planetary systems that may have dilithium crystals." In seconds, the computer complies.

Don't we all wish that we could have computers activated by voice commands? That day may arrive sooner than you think.


Get all the Monitor's headlines by e-mail.
Subscribe for free.

Several years ago, I became intrigued by the possibilities of voice-recognition software. For one thing, I'm a lousy typist. So anything that would improve my accuracy was more than welcome.

The problem was that voice-recognition programs built five or six years ago made the problem worse. The best programs offered 95 percent accuracy. But only someone who had been taking elocution lessons since kindergarten could hope for such results. When I tried the old programs, I spent as much time making corrections as I did in actual dictation.

Then about two weeks ago, I received an e-mail from Nuance, a software company, that invited me to try out the latest version of its Dragon NaturallySpeaking software. And I'm glad I did.

Version 9 of Dragon NaturallySpeaking promises 99 percent accuracy. While that may be optimistic, the results that I've had are pretty darn close. Clearly, it is a big leap forward from past experiences.

But what's even more interesting are those possibilities hinted at earlier. Voice-recognition software has always been a boon to disabled folks who have difficulty typing. But it hasn't been much help to those who can type because of the amount of time required to properly "train" the software to understand your voice. Not to mention the patience needed to make the endless corrections.

But the new levels of accuracy possible with the new software open up a world of different tasks. For instance, I've been using it to answer my e-mails each morning. Not only do I answer them faster, but I find that I answer more of them. (I confess I put a few e-mails aside each day with the thought, "Well, I'll answer them later," which, of course, I never do.)

Nuance already makes versions of Dragon NaturallySpeaking for the legal and medical professions, but now I can see college professors, business people, police officers - basically anyone who writes a lot of reports - finding the software enormously useful.

I'm also sending a copy to my mother, who spends hours writing e-mails to people across America and in the British Isles, searching for information about my family's history. Since Version 9 requires no training at all, it's easy for nontechie people like her to start using it immediately. (And I don't have to spend hours on the phone helping her make it work.)

And then there's my profession, journalism, and in particular, bloggers. (Yes, I consider many bloggers to be journalists.) One of the keys to blogging is writing in your own voice. Well, now you really can write in "your own voice." And it's my bet that bloggers who use the software will post more material because it's just so much fun to use.

But before you rush out to buy Dragon NaturallySpeaking there are a few caveats. Good voice-recognition programs take up a lot of space on your computer. It requires about 512 MB of RAM and at least 1 GB of hard-drive space. So those with older computers may need to update their machines. In addition, there is no version for Apple computers. (David Pogue of The New York Times says if you own a new dual-boot Apple that runs Windows, you should give Dragon NaturallySpeaking a try.) The only other concern is that it's still difficult to use in a noisy environment (like my house that's filled with four children).

Dragon NaturallySpeaking sells for about $75. A "preferred" version will set you back $150. But I must say that voice-recognition software has finally reached a point where it is useful to everyone. For instance, I dictated this column and made a total of six corrections. Only in one case did the program fail to do what I wanted it to do and I had to intervene to fix it manually. Pretty good.

Besides, any program that spells "dilithium crystals" correctly is absolutely worth a try.

To learn more about voice-recognition programs check out http://www.consumersearch.com/www/software/voice-recognition-software/index.html.

Get Monitor stories by e-mail:
(Your e-mail address will be protected by csmonitor.com's tough privacy policy.)

In Pictures
Fireworks: A party in the sky

ELECTION '08 Patchwork Nation
The American voter beyond red and blue

FISHERIES Empty Oceans Series
The sea is no longer so vast.


Daily podcast

Monitor Reports

Discussions with Monitor reporters from around the world


Today

Peter Grier

Honduras has two presidents, but no solution to the country's political crisis.




Making a difference
Making a Difference

What happens when ordinary people decide to pay it forward? Extraordinary change. See how individuals are making a difference, finding solutions, overcoming adversity, and giving back globally.

Jeremy Gilley, founder of the nonprofit Peace One Day, talks with students at Cambridge Rindge and Latin High School in Cambridge, Mass.

Melanie Stetson Freeman/Staff

People making a difference: Jeremy Gilley

This actor and filmmaker envisions that world peace begins with just one day of peace.