Path: utzoo!attcan!uunet!clyde.concordia.ca!news-server.csri.toronto.edu!cs.utexas.edu!yale!bunker!hcap!hnews!385!14!Wayne.Roorda
From: Wayne.Roorda@f14.n385.z1.fidonet.org (Wayne Roorda)
Newsgroups: misc.handicap
Subject: Voice Recognition
Message-ID: <14814@bunker.UUCP>
Date: 9 Oct 90 13:02:28 GMT
Sender: wtm@bunker.UUCP
Reply-To: Wayne.Roorda@f14.n385.z1.fidonet.org
Distribution: misc
Organization: FidoNet node 1:385/14 - Fingers Talk, Lawton OK
Lines: 37
Approved: wtm@bunker.UUCP

Index Number: 10961

In this past week's(October 1, 1990) PC WEEK newspaper on page 5 there
is an article titled "Soliloquy Provides the Latest Word on Voice
Recognition" by Frederic E. Davis.

The table which I am copying describes the differences between current
technology vs. Soliloquy's technology.

Current Speech-Recognition              Soliloquy's New Speech
Technology                              Recognition Technology
-------------------------------         ----------------------------

Is based on how people listen           Is based on how people talk

Uses a recognition algorithm            Uses a recognition algorithm
based on spectral pattern               based on mouth movement
matching

Recognizes a rigid grammar that         Recognizes the grammar of
predetermines sentence structure        natural speech and
                                        transpositions

Requires large amounts of RAM           Requires limited RAM

The software was developed by Emerson & Stern Associates, Inc. and
was announced at the American Voice Input/Output Society forum in
Bethesda, Maryland.  Soliloquy was written in "C" and is reported
to be generic enough to port to DOS, Unix and OS/2 systems.

If Soliloquy is all that it claims to be should be a real assist
for all of us.  Comments anyone?
                                            Wayne

--
Uucp: ..!{decvax,oliveb}!bunker!hcap!hnews!385!14!Wayne.Roorda
Internet: Wayne.Roorda@f14.n385.z1.fidonet.org