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 experiences please From:soundy -at- NEXTLEVEL -dot- COM Date:Sun, 18 Feb 1996 14:06:27 -0600
In <1996FEB17 -dot- 2904 -at- cerebus -dot- ozemail -dot- com -dot- au>, on 02/17/96 at 08:44 AM,
SANDRA CHARKER <scharker -at- ozemail -dot- com -dot- au> said:
>Is anyone out there dictating their deathless prose instead of typing it?
I sure wish! I've seen IBM's VoiceType Dictation in action - this is the
kind of stuff you see in James Bond movies (heh - just saw Goldeneye at
the $2 theatre last night).
>I'm getting a new notebook computer, with built-in microphone. Is anyone
>using voice recognition software? How do you like it, and how much do you
>do with it? Is it impossibly expensive, or just outrageously?
Check your local stores for the top-end IBM Aptivas - the ones that come
with OS/2 Warp preloaded also have VoiceType included - give it a try.
At this point, I believe VoiceType is limited to a proprietary audio-input
card, or the Mwave DSP-based sound system used in the Aptivas; however, if
you're looking for a laptop, consider an IBM ThinkPad, which uses the same
Mwave chipset as the Aptivas - I'd have to check, but VoiceType should
work equally well with that.
The demonstrations I've seen of it were amazing. It does a good job at
*speech* (as opposed to voice) recognition right out of the box, and with
one of a few simple thind-party add-ons can provide voice control of your
computer. With an hour or two's training to your voice, it will easily do
90+ words per minute of speech-to-text, and unlike most voice-recognition
software, it won't care if you have a sore throat or plugged-up nose -
it's primarily SPEECH recognition.
It also has a built-in, on-the-fly spelling and grammar checker that will
very intelligently go back and make context corrections as you continue to
dictate. For example, if you say "two", it may enter "to" until you get a
few words farther down, it realizes you meant the number, and jumps back
to replace the "to" with "two".
The demos I've seen ran about an 85-90% accuracy rate, and the system had
not been fully trained. It claims better than 95% accuracy.
Your friend and mine,
Matt
----------------------------------------------------------------------------
---*** NOTE: Windows 95 Fixpack now available! ***---
See http://www.austin.ibm.com/pspinfo/drk95.html for more info!
Opinions expressed do not necessarily reflect those of Next Level
Productions, or anyone else of sound mind from this planet or
dimension!