Newsgroups: comp.speech
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!cs.utexas.edu!asuvax!ennews!trcsun3!deisher
From: deisher@trcsun3.eas.asu.edu (Michael E. Deisher)
Subject: Re: Recog. Accuracy; IN3, CSS, CVA, etc.
Message-ID: <D28yz5.BEp@ennews.eas.asu.edu>
Nntp-Posting-Host: enws125.eas.asu.edu
Sender: news@ennews.eas.asu.edu (USENET News System)
Summary: please define "moderately poor"
Organization: Arizona State University
Date: Wed, 11 Jan 1995 15:28:15 GMT
References: <3eetbj$1vah@tequesta.gate.net> <3eu4ke$jq@newsbf02.news.aol.com>
Lines: 17

In article <3eu4ke$jq@newsbf02.news.aol.com>, jean1112@aol.com (Jean1112) writes:
> >Can anyone advise their comparative experiences on the accuracy
> >(or lack of it) between the various Windows-based voice recognition
> >packages?
> 
> Chris, please contact me at vinces@vnet.ibm.com for information on using
> the ICSS recognizer.  We provide good speaker independece, and may be able
> to help with your project.  Particularly, if you are willing to desgn the
> phase structure carefully we have been able to demonstrate good
> performance under moderately poor signal-to-noise conditions.

Vince, would you please clarify what you mean by "moderately poor"?

Thanks,

--Mike
deisher@dspsun.eas.asu.edu
