My favorite productivity tool: voice recognition software

Next week, Blawg Review, The Carnival of Law Bloggers, will be hosted at Legal Andrew: Productivity ideas for the legal world: law students, lawyers, Westlaw, and Lexis.

Not surprisingly, the anonymous editor of Blawg Review hints that “legal productivity” posts will be highlighted.

Thinking about my experience with that subject, the first thing that comes to mind is that blogging and reading blogs may on occasion help enhance legal productivity, but too often in my case hurts it.

The second thing that comes to mind is that this is a perfect opportunity to promote my favorite productivity tool, voice recognition, specifically Dragon Naturally Speaking, by Nuance.

I’ve been using this product for many years — too many to count. They’re now on version 9 and I know I’ve been using it at least since version 6.

During that time, periodically I’ve read negative reviews of voice recognition, generally to the effect that the required training is too much of a hassle and that accuracy is disappointing.

I was always puzzled reading such remarks, as I had almost immediately become addicted to using voice recognition for all of my written work. And even if there was some merit to these complaints with earlier versions, the current version rocks.

At this point, the required training is minimal to nonexistent. Speed and accuracy is astonishing. I’ve clocked it at over 130 wpm using online typing tests, taking errors into account. And the price is a steal. I use the “preferred” version, not the considerably more expensive legal version. My use of fancy legal vocabulary is limited enough that I have no need for the legal version; any special vocabulary can be readily incorporated using very efficient customizing options.

How does voice-recognition add to my productivity?

I was never a good typist. I had difficulty learning to dictate in a manner that produced results that did not require extensive editing to suit my perfectionistic writing style. With voice recognition, I have the benefit of extremely rapid typing, combined with the ability to compose and edit on-screen as I go — as opposed to dictating a tape, unable to efficiently review and edit until after I’ve waited for someone else to type it.

What does voice-recognition offer if you are a fast typist happy with your ability to compose on-screen as you type?

You will benefit from even greater speed. You will do your body a favor, reducing the risk of repetitive motion injury. I do a tremendous amount of Web surfing for this blog, which involves lots of mouse clicking. I feel some wrist strain from this. I just discovered that with Naturally Speaking I can just point the mouse and say “mouseclick.” Do that a few hundred times a day and think of all the ligament strain you’re avoiding.

This post was dictated using Dragon NaturallySpeaking voice recognition software


  1. A couple of caveats on further reflection:

    This program is a memory hog. I have a Gig of RAM on my laptop, and it still works better if I do not have too many other programs open.

    It performs best with a particular speaking style. It is designed for continuous speech recognition, which means it does better if you speak in clauses rather than single words. It uses the context to assist in recognizing each individual word.

    However, this does not mean you can run your words together as you might in casual conversation and get excellent results. For best results, you need to enunciate each word carefully, even if speaking at a high speed.

  2. I think that voice and speech recognition software will transform the way we interact with computers. Instead of a one-way interaction (i.e., you typing and clicking information into the machine), the use of computers in the near future will involve two-way communication. It’s a higher level of interaction with the machine.

    I have a video demonstration on YouTube:

    As you can see, there’s a whole world of audible computing technology.

    I’ve been using my voice recognition software on an IBM Thinkpad 800MHz laptop that has only 256M ram and only a 20Gig HD. I use the Speech Engine from Microsoft.

    Very cool stuff…

Leave a Reply