1) Yes, voice recognition (VR) is hard. Perhaps the most difficult thing we
have so far managed to get PCs to do at all well. So the present,
regretable state of the art is nonetheless awesome. (Think for a moment
about the difficulty people have in doing good voice recognition. Small
children take years to learn this skill. Grownups often have an even harder
time learning to do it in a new language. And one makes lots of mistakes
when listening to a speaker with an odd accent.)
2) There are alternatives, though some of the most interesting ones are not
yet "ready for prime time." To my way of thinking, the ultimate PC input
device is one that senses what one intends and acts on that directly. [This
would make moot the joke that reminds one that (now) PCs do only what we
tell them to do, instead of doing what we want them to do.]
Does this mean a mind reading PC? Well, almost, but not quite. If one can
sense the minute electrical signals leaking out from the brain, or pick up
the larger and possibly related muscle-activating signals, and sense these
signals in the skin surface, then in principle at least it would be
possible to "tell" our PCs anything we might wish merely by thinking of
those things. We'd need to train both ourselves and our PCs to accomplish
this feat. And make no mistake: This recognition process and the required
personal training are likely to be even harder than those involved in VR.
However, to let you know that there is hope along these lines, check out
the following URL:
I asked and yes, these people are using both muscular signals and brain
wave signals. They claim at present that most users are only able to train
themselves to produce about a dozen sufficiently distinct signals so they
are only now able to let one choose one of twelve possibilities. But that
situation will surely improve with additional research.
I don't know if this technology is within the budget of any of the readers
on this list (yet), but just knowing that people are working on it may give
you hope and another reason to persist in your present efforts until such
time as this new approach is ready for you.
John
At 03:05 PM 5/26/98 -0400, you wrote:
>
>Well many thanks to all of those who responded to my long-winded tales of
>woe. I'd like to summarize the advice that I was given, in case it's
>helpful to others, and many of the responses didn't get posted. I guess
>people wrote to me directly.
...{most of content removed]...
>So now I am whining and complaining. I do think voice recognition is a
>wonderful technology, and it has certainly come a long way, but I still
>don't think it's ready for prime time yet. Even with the best of systems,
>the user still has to be a disciplined, patient, techno-geek with a lot of
>time on their hands. I myself am only a modest techno-geek, and rate poorly
>in the other categories. Maybe if I only had to use it once in awhile I
>could handle it, but I don't think I can put up with this kind of
>frustration all day long.
=======
John M. Goodman, Ph.D., author of "Peter Norton's Inside the PC,"
Seventh Edition (Sams 1997, ISBN 0-672-31041-4)
=======
![]() |