Disclaimer

The content of this blog is my personal opinion only. Although I am an employee - currently of Nvidia, in the past of other companies such as Iagination Technologies, MIPS, Intellectual Ventures, Intel, AMD, Motorola, and Gould - I reveal this only so that the reader may account for any possible bias I may have towards my employer's products. The statements I make here in no way represent my employer's position, nor am I authorized to speak on behalf of my employer. In fact, this posting may not even represent my personal opinion, since occasionally I play devil's advocate.

See http://docs.google.com/View?id=dcxddbtr_23cg5thdfj for photo credits.

Sunday, August 02, 2009

Dragon Speech Recognition

I have just installed Dragon speech recognition. I used to use Dragon speech recognition more than 10 years ago, but gave up, not because of speech recognition quality, but because it annoyed cubicle desk neighbors, and mainly because it became a pain to have to reinstall on every new computer.

I have opened this blog entry to record a few impressions as I start using Dragon again after this decade of nonuse.

I think that I needed to enter in more training this time around about 12 minutes of reading Arthur C. Clarke.

I thought that under the old Dragon I was able to turn the microphone on and off by voice. Or, rather, I think there were three microphone modes: (A.) completely off can't be turned on or off at all by voice; (B.) completely on, used for all commands; and (C.) a mode where the microphone was actually on but where all commands except something like "turn the microphone on" were ignored. The last mode was convenient for disabling accidental voice commands.
Ahhh... there is such a mode: "go to sleep" or "stop listening."

The Dragon bar has something called a select and say indicator, that is supposed to indicate whether the application you are connected to supports all of Dragon's functionality. I find it rather amusing that the application bar is yellow when connected to Dragon naturally speaking tutorial. I.e. the tutorial program doesn't support Dragon completely.

It is a bit sad that Firefox's text window, such as I am speaking this blog entry into, inspires Dragon to say "dictating to a nonstandard window". I wonder if this will be enough to force me to use Internet Explorer for my blog entries.

It is already evident that speech recognition, once again, allows me to be more verbose than when I was typing. Since I am already pretty verbose even when I am typing, this may be considered a downside. On the other hand, it looks like Dragon's spelling is better than my spelling when I am typing.

Dragons undo and redo capabilities seem somewhat restricted. Particularly when you are typing in words like "undo" and "redo". I think the biggest annoyance that I've had with speech recognition so far is accidentally undoing.

Microsoft's habit of automatically changing the focus to Windows as they pop up just caused me a problem: I started speaking into a pop up window.

Dragon's correction occasionally has problems: I just tried to correct "cause" into "caused", and Dragon recognized the correction correctly, but somehow got the insertion of the correction incorrect, resulting in "ccaused".