Newsgroups: sci.engr.control,sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!udel!gatech!swiss.ans.net!news.dfn.de!gs.dfn.de!gwdu03.gwdg.de!rschirm
From: rschirm@gwdu03.gwdg.de (Rolf Schirmacher )
Subject: Re: Questions on Basic Signal Processing
Message-ID: <57FUBYZV@gwdu03.gwdg.de>
Followup-To: sci.engr.control,sci.image.processing
Sender: news@gwdu03.gwdg.de (USENET News System)
Nntp-Posting-Host: gwdu20-eth.gwdg.de
Organization: GWDG, Goettingen
X-Newsreader: TIN [version 1.1 PL8]
References: <3a8gk4$kct@girtab.usc.edu>
Date: Tue, 15 Nov 1994 10:51:13 GMT
Lines: 29
Xref: glinda.oz.cs.cmu.edu sci.engr.control:3486 sci.image.processing:10892

As I am not affiliated with Sony (or any other company producing such
CD players) I can only guess and tell what I would do.

The most easy way seems to be to record "typical" impulse responces 
from different kind of rooms, put them to a digital filter and put the
(switchable) filter in front of the D/A-converter. If you take an IIR
filter (one with poles), it is possible to model quite long reverberances
with a low number of filter parameters / a low numerical complexity. 

This implies that the CD itself has been recorded echo free (e.g. in an
anechoic chamber or convoluted with the inverse impulse responce of the 
recording room), but in practice this exactness is perhaps not needed
(you won't get the exact MET sound but somthing that sonds like a big opera
hall...).

But perhaps somone on comp.dsp or sci.physics.acoustics can help further?

Rolf


--
------------------------------------------------------------------------------
Rolf Schirmacher			Tel. 	+49 551 39 7733
III. Physikalisches Institut		e-mail	rolf@schirm.physik3.gwdg.de
Universit"at G"ottingen				rolf@physik3.gwdg.de
B"urgerstra"se 42 - 44				rschirm@gwdg.de
D - 37073 G"ottingen
------------------------------------------------------------------------------

