Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!caen!spool.mu.edu!olivea!oliveb!bunker!hcap!hnews!382!4.0!Arthur.Clarke
From: Arthur.Clarke@p0.f4.n382.z1.fidonet.org (Arthur Clarke)
Newsgroups: misc.handicap
Subject: Re: IBM voice recognition
Message-ID: <16329@handicap.news>
Date: 21 Jun 91 13:57:17 GMT
Article-I.D.: handicap.16329
Sender: wtm@bunker.isc-br.com
Reply-To: Arthur.Clarke@p0.f4.n382.z1.fidonet.org
Organization: FidoNet node 1:382/4.0 - River City MailBox, Austin TX
Lines: 25
Approved: wtm@bunker.hcap.fidonet.org

Index Number: 16329

Thanks for the practical note.  Yes, you do detect frustration
at the almost unuseably slow word per minute thruput and the
required artificial pause between words that sounds like it will
shatter the concentration needed to compose quality text.
    However, it is early in the technology and fixes are likely
to be found.  It will be interesting to evaluate the various
products.  Since Austin has large IBM, Motorola (read source of
Apple chips) and UT (as a Cray in the basement at @ 2am if need
be) resource bases plus enough DC connections to ease Defense
Dept and black patent problems in case that has watered down
released capability, I would expect that applied effort will
result in improvement.
    Not all that much fix is needed.  Something like 60-70 WPM
and pause patterns more like when you stop for breath anyway
should be doable be some means or another.
    Evaluation and user comment is the next step.  If you or
anyone else would like to boost the process along, just drop a
reply with your MAIL addresh your MAIL address for obvious
reasons.

--
Uucp: ..!{decvax,oliveb}!bunker!hcap!hnews!382!4.0!Arthur.Clarke
Internet: Arthur.Clarke@p0.f4.n382.z1.fidonet.org