Wired has a couple articles up today about the state of voice recognition software (Voice Recognition: Still Trying and When Your Voice Is All You Have), that reaches pretty much the same conclusion I have from personal experience — today’s software delivers, the problem is getting used to dictation as a way of working.
Several months ago I bought one of those combination headphone/microphone sold by Plantronics and installed the latest version of Dragon Naturally Speaking on my PC. After about 15 minutes of training the program to recognize my voice I was up and running. I played with it for a few weeks and was very impressed by how accurate the software was. In my opinion the Dragon software is already mature enough right now to be a complete dictating solution for somebody with repetitive stress injuries.
On the other hand, I haven’t used the software at all after the initial few weeks. Why? Because dictating, as opposed to writing or typing, requires a very different mindset that I just don’t have time to adapt to. I’m apparently not alone, as Wired paraphrases computer science professor Ben Schneiderman saying,
Talking at your computer significantly interferes with your thought process, he says, because thinking and speaking are both drains on a person’s short-term and working memory. Physical activity, on the other hand, taxes another area of the brain altogether. That’s why it’s much easier to talk while you walk than it is to speak while you think.
One of the writers profiled by Wired who does use dictation software said it took her several months to get used to dictating as opposed to typing.
It’s good to know that voice recognition technology could replace typing if I suffered from a severe RSI injury, but otherwise the time investment required to get used to dictating is just too high to make the software useful on a regular basis.