TechWhirl (TECHWR-L) is a resource for technical writing and technical communications professionals of all experience levels and in all industries to share their experiences and acquire information.
For two decades, technical communicators have turned to TechWhirl to ask and answer questions about the always-changing world of technical communications, such as tools, skills, career paths, methodologies, and emerging industries. The TechWhirl Archives and magazine, created for, by and about technical writers, offer a wealth of knowledge to everyone with an interest in any aspect of technical communications.
Subject:Re: voice recognition (cont'd) and IBM From:P Newman <pnewman1 -at- HOME -dot- COM> Date:Sat, 7 Aug 1999 06:00:09 -0400
Hi Robert,
From your brief description of your error correction, it sounds to
me as if you mgmt not be teaching the program. When you are
correcting, how are you correcting? Please give a step by step
approach. Do not feel you are giving too many details. If you are
simply editing the errors, the program will not learn. I am shooting
from the hip on this since I have only worked with NatruallySpeaking
and L & H, but the concept in all programs is that they learn from
following the proper correction procedure and saving your voice file
with the changes. A simple edit of the error does not teach the
program. You must use the built in correction facility. Please tell us
exactly what you are doing and we will try to help.
Peter
----- Original Message -----
From: ROBERT SIDMAN <scribe -at- EZONLINE -dot- COM>
To: <TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU>
Sent: Friday, August 06, 1999 5:43 PM
Subject: voice recognition (cont'd) and IBM
> I have been following the thread on voice recognition systems. As
one who
> depends on a system, I thought I would get my $.02, maybe more,
in. Since
> my December, 1997 stroke, and more particularly since May, 1998,
when I
> returned home, I have depended upon the IBM voice recognition
software to
> communicate.
>
> My left arm is now is useless. Typing with only my right hand is a
slow,
> laborious process. Dictating to my voice recognition software is
slightly
> faster, but we may be comparing the racing speeds of turtles and
snails. I
> hope some IBM types are lurking and will read this note.
>
> I originally was using IBM ViaVoice. After upgrading my CPU at home
to a
> Pentium II chip and going to Win98 and 128 megs of RAM, my
(federal) employer
> upgraded me to ViaVoice Gold, which is also installed on my office
computer.
>
> It may be me, but there are many times when I use the system it
sucks, big
> time! There are fewer times when it works well. Apparently I find I
must stop
> dictating every 3-4 sentences, edit and correct the copy, save it,
and
> continue.
>
> Even though I have been used the system daily for more than six
months, I am
> still amazed at the amount of mistakes it makes from my dictation.
The
> learning process, particularly in teaching the computer to recognize
my speech
> pattern and vocabulary, will never end. This is a major
shortcoming.
>
> Another major shortcoming is that to write this message, I must
first dictate
> to MS Word or Notepad. When the note is already to be sent, I must
copy and
> paste it to Eudora Pro software. There is no way I can directly
dictate to
> Eudora, or Hot Dog or Arachnophilia. This is very frustrating. I
have also had
> problems trying to dictate using Lotus Notes, another (fine?) IBM
product. If I
> was not now extremely folically challenged, I would be for using
this package.
>
> On a 1-10 scale, I'd rate it as a 3-4. It does not have a good
beat, it is
> inconsistent in performance, and you definitely cannot dance to it.
>
>
======================================================================
=====
> Send commands to listserv -at- listserv -dot- okstate -dot- edu (e.g., SIGNOFF
TECHWR-L)
>
>
>
>
