Newsgroups: comp.speech
Path: lyra.csx.cam.ac.uk!warwick!pipex!howland.reston.ans.net!agate!msuinfo!uchinews!cdsmail!timbuk.cray.com!walter.cray.com!hydra!btalbert
From: btalbert@hydra.cray.com (Bob Talbert)
Subject: Re: Tomorrows World Speech Recogniser
Message-ID: <1994Apr27.010829.14200@walter.cray.com>
Lines: 10
Nntp-Posting-Host: hydra.cray.com
Organization: Cray Research, Inc.
References: <1994Apr19.111807.8191@leeds.ac.uk> <jimn8CoIv7D.JJ@netcom.com> <berry.debruin.65.2DBBB2BC@mi.rulimburg.nl>
Date: 27 Apr 94 01:08:26 CDT

  I saw the program. They were useing the Sony portable DAT machine.
They could have recorded the voice @ 44KHz or 32KHz (or is it 38KHz).
  The speakers put out a lot of bass. This may be important. The speakers
were small self powered units. 

  Very impressive report! Background noise improved the correct/acceptible
% from 78% to 89% when the lady talked into the microphone. The DAT recording
showed only about 5%. Sound too good to be true? Was the recording doctored
or was the freq. responce of the speakers modified? Who knows.

