Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!news.duke.edu!convex!cs.utexas.edu!math.ohio-state.edu!usc!elroy.jpl.nasa.gov!decwrl!amd!netcomsv!netcom.com!daphne
From: daphne@netcom.com (Daphne Gould)
Subject: Re: REQUEST: PCMCIA based speech recognition systems
Message-ID: <daphneCwv1zC.J02@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <1994Sep23.170918.19609@dcdwest.uucp> <362o2p$6ni@rigel.infinet.com> <366kr7$hli@gatekeeper.ddp.state.me.us> <1994Sep27.164940.11902@dcdwest.uucp>
Date: Wed, 28 Sep 1994 22:18:47 GMT
Lines: 23

In article <1994Sep27.164940.11902@dcdwest.uucp>,
Peter Benson <benson@dcdwest.uucp> wrote:
>hwjsteu@gatekeeper.ddp.state.me.us (John Steuerwalt) writes:
>
> ...
>
>What I found out yesterday is that Dragon Dictate and other Dragon
>System products employ software only solutions to speech recognition.
>They can use a PCMCIA card for digitization.  No PCMCIA card I was able
>to uncover did any form of digital signal processing for speech
>recognition (other than analog to digital conversion).  That is, there
>does not seem to be a plug in PCMCIA card that recognizes speech.
>
>BTW, I would really like my findings to be false. If anyone knows
>better, please tell me.

Concerning Dragon products, your information is basically correct.  
DragonDictate for Windows will use a PCMCIA card as its analog to digital 
converter and then DragonDictate will perform both the signal processing 
and the recognition algotithms in the host processor (i.e. the '486).

Joel Gould
Dragon Systems
