Several years ago I gave voice recognition a try. It looked promising and efficient (and a friend raved about it, but I think that’s because he owned stock in one of the companies). So I did some research, bought one of the more popular versions and installed it on my Windows 98 machine (yee-hah!).
I donned the headset microphone, fired it up, and went through the “teaching” phase, where the software displayed five hundred quotes, famous or otherwise, and I recited them back so that it could learn my accent. That took maybe an hour or two. I also glanced through the tutorial and realized that not only could I dictate the words, but also the editing, i.e. if I paused, then told it, “delete the last four words”, the software would suspend dictation, pick up on the word “delete”, and then delete the last four words I had typed. On the downside, I also had to dictate punctuation, which isn’t a big deal if I were writing a memo, but got tricky during dialog. For instance, to get
“Don’t you think the software might have improved in the last decade?” Matt blurted impatiently.
I would have to say out loud, “Indent open quote don’t you think the software might have improved in the last decade question mark close quote Matt blurted impatiently period.”
…which is a little jarring, but I suppose with a lot of practice I might get used to it.
So I started. I said, “Dante walked into the alley, suspicious of the guy who was following him, period.” This appeared on the page:
Dante walked into the alley suspicious of the guy who was following him.
I wanted to change that though, so I paused and tried the editing commands.
“Insert a comma after alley.”
Dante walked into the alley suspicious of the guy who was following him. Insert a comment after alley
“No, Delete that last sentence.”
Dante walked into the alley suspicious of the guy who was following him. Insert a comment after alley no delete that last sentence
“No no no,” I said, ignoring the obvious.
Dante walked into the alley suspicious of the guy who was following him. Insert a comment after alley no delete that last sentence no no no
“You fucking suck,” I blurted, frustrated.
Buck a buck
And….that ended my brief experiment with voice-recognized writing. Perhaps with a lot of patient teaching and editing, I could get the software to do what I wanted, but I quickly realized it was just easier for me to touch type.
Granted, this was a decade ago, and since then voice recognition has come a long way. Although I’m very comfortable with my current methods (typing allows me to express the forming thought at a comfortable pace), I might revisit it. I can see its advantage in the legal profession, because among other things, the software let you set up templates and hotkeys, where you could say “Insert standard paragraph 1-B”, and it would just dump in predefined, proofread, boilerplate text. So you could verbally drag-n-drop a complete, complex legal document in a few minutes, and then search-replace the relevant terms.
It’s a little harder with novel writing, I imagine, and probably wouldn’t work in a cube farm because of the noise, but are other folks using it? What are your experiences?