Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!swiss.ans.net!newsgate.watson.ibm.com!watnews.watson.ibm.com!bocanews.bocaraton.ibm.com!news
From: sumner@vnet.ibm.com (Charlie Sumner)
Subject: Re: Speech Recognition
Sender: news@bocanews.bocaraton.ibm.com (News Admin ID)
Message-ID: <Cz0t64.9I7@bocanews.bocaraton.ibm.com>
Date: Wed, 9 Nov 1994 22:01:16 GMT
Lines: 21
Reply-To: sumner@vnet.ibm.com (Charlie Sumner)
References: <39r2ng$io8@news.kth.se>
Nntp-Posting-Host: sumner.bocaraton.ibm.com
Organization: IBM, Boca Raton, FL
X-Newsreader: IBM NewsReader/2 v1.00

In <39r2ng$io8@news.kth.se>, pj@linnea-grind.stacken.kth.se (Patrik Johansson) writes:
>Can someone please provide me with information on different speechrecognition
>solutions to PC? I know you have been discussing this but i  would need a 
>"wider" description.
>
All of the systems are listed in the FAQ.  There are many solutions, but you need
to be a little 'narrower' in your request.  There are speaker independent/speaker dependent
systems, dictation vs. navigation, continuous vs. discrete, MAC, Windows, UNIX, and OS/2 systems.

>So, what systems are there, and what does they require in software
>and hardware? The only macs today that can handle speech recognition has
>build in DSP, is the same required on PC?
>
That really depends on the above.  Most continuous speech navigation systems don't require a 
DSP since the vocabulary is limited and the processing does not overload the current 
processor capabilities (486 and the like).  If you get into dictation with large 
vocabularies, then most systems either require, or make use of, a DSP if it is available
to speed up recognition.

Charlie Sumner   IBM Corp.  Boca Raton
sumner@vnet.ibm.com

