I like to say that I’m not a gadget guy, but this post may prove me wrong.

While shopping for other, useful things, I picked up a copy of some speech recognition software. My notebook computer’s a lot more powerful than the ones I’ve had in the past, so I thought it might be easier this time to learn how to boss the computer around with just my voice.

And actually, it’s working pretty well.

I began by sounding like a computer-generated voice — too slow, no affect. The more I practice, even if I slow myself down by dictating e-mail instead of typing it, the fewer pauses I tend to make, and the more the software and I seem to get along.

Dictating instead of writing doesn’t make much sense for short bursts, and certainly doesn’t seem to make sense for filling in forms. I’m not sure I want to learn how to dictate all the key commands as well as the words or letters that I want to enter.

On the other hand, I do a lot of extended writing. (Some people might say far too much, but that’s a topic for another day. ) I thought it would be worth experimenting with speech software, both to see whether it change my writing style, and to avoid some of the longer stretches of typing that I sometimes find myself in.

As part of my practicing, I’ve sent e-mails to a few people. Having sent them, I’ve learned I need to proofread more carefully, because while speech recognition clearly has gotten better, What You Say Ain’t Necessarily What You Get.

The software is supposed to work easily with standard word processing and e-mail packages. For other applications, like this blog, there’s a special dictation box. That’s what I’m using for this post.

I tried dictating text directly into the blog, but there was a lot of lag, probably due in part to the autosave feature that’s part of my blog software.

With the dictation box, the excessive lag went away (leaving only “normal” lag). When I’m done, there’s a transfer button to move the text into the actual application.

I once heard Jim Fuller say, “Practice doesn’t make perfect — practice makes permanent.” His point was the need for specific, useful feedback, without which you end up entrenching poor practice.

(That’s a less-than-obvious outcome of what I’ll call uninformed practice.)

For the most part, I’m getting that specific, prompt feedback with this program.

I do see a certain lag between when I say the words and when they appear on screen. That can be disconcerting, like when you’re on an international phone call, and there’s that quarter-second lag as the sounds bounce off the satellite.

As with the long-distance calls, I imagine you learn to deal with the software’s lag.

For some reason, the command I have the most trouble saying (or, to be precise, the command I have the most trouble getting obeyed) is “go to bottom,” which is supposed to move the cursor to the end of the document. I’m not sure why, but so far I haven’t mastered that. As a result, some of the simple editing that you theoretically can do is a problem for me.

I’ll keep trying, though. I’m a pretty fast typist, but I know for a fact I can talk faster than I can write.