A long time ago I bought one of the first versions of Dragon Naturally Speaking. Or whatever it was called at the time. This was 1997. I remember trying it out and being frustrated by the fact I had to train it to understand my speech. Fast forward to 2010 when I got my first android device and started using voice to text conversion. I was always a little disappointed with the way Google’s conversion worked but it worked well enough for what I was using it to do. Then I replaced my android device with an iPhone 4S with Siri. I instantly fell in love with the Siri voice recognition, even though it occasionally has major meltdowns. Its ability to understand natural speech is wonderful and helps me take down notes and thoughts which I then front to other task managers.
I understand Dragon NaturallySpeaking works exceptionally well nowadays. I know there are full-time authors who use it exclusively to write their books. David Weber is an example of this. With Griffin in the house, I find much less time to sit down and draft blog posts. I do not have two hands free even when he’s sitting there next to me just being a squirmy baby. As everyone keeps telling me, having a baby changes everything. Including, I guess, how I’m going to write on my blog.
Unfortunately, using voice recognition is going to introduce a competency problem.
I’ve never had the opportunity, or occasion, to do a lot of dictation like you see in old movies with the boss dictating letters to the secretary. I have grown up in a situation where all of my first drafts are done with a word processor, and the final draft is done by modifying that draft using cut and paste and other tools that came about with the advent of the computer. I have learned through experience how to draft a document by thinking and typing; by editing sentences on the fly and by cutting and pasting as necessary. Moving whole paragraphs around is something I’m used to. I’m not used to stream-of-consciousness composing from beginning to end.
I realize that the dictation/voice recognition element is merely the first step in composing a document, with editing and rewriting to follow, I still find it a bit intimidating to try. For example, about half of this posting was done through dictation to Siri. You may be able to tell which half. I certainly can. Siri, while a good voice translator, does not allow for the long-winded sentences that I like to use. You have to stop occasionally so Siri can do the translation. The PC-based systems don’t have that problem, but there’s a usability issue: I’m not on my PC very much anymore. It’s a hassle to open it and boot up merely to do a blog posting. A PC-based voice recognition software would be useful to me if I were writing longer documents, but knowing myself, I think I’ll probably be doing a lot of Siri-dictation in the near future.
Given that I also own a Chromebook, I am interested in a web-based application that would allow me to do voice-to-text conversion. This would allow me the “Hey, I need to write this down NOW!” ability that turning on my PC does not. Unfortunately, it doesn’t look like there are any available that have the features that Dragon does1. The Google voice recognition seems to be only for Google searches. So looks like I’ll be using Siri, or nothing. Or, it may be this is all unnecessary rambling as Griffin may change his habits to allow for more writing time for me. We’ll see.