room reverberation simulations (Jont Allen )

Subject: room reverberation simulations
From:    Jont Allen  <jba(at)>
Date:    Thu, 28 Feb 2002 11:07:17 -0500

> > > Date: Tue, 26 Feb 2002 22:36:56 -0800 > From: John Hershey <jhershey(at)COGSCI.UCSD.EDU> > Subject: simulated reverberation > > What is the most natural way to computationally simulate realistic > reverberation? > I am told that real reverb is non-linear, due to the moving around of > air, so you in principle you can't just run it through a linear filter It is true that in a large room that the impulse response will vary with time, due to air currents, fluctuations in temperature, and the like. However this is not a nonlinear effect, it is rather a time-varying effect. To be nonlinear, the signal would need to interact with the impulse response. That is not an effect one must worry about normally. The sound in a horn loudspeaker might become nonlinear, but not the room impulse resp. I wouldn't worry about the time varying effect either, unless you wish to attempt to invert the impulse response, or you want to simulate an auditorium. There is a matlab version of my room image program on the net. I cannot attest to its accuracy (The JASA fortran prog has no known bugs, that I know of) Jont Look at: (at)article{Allen79b, author = {Allen, J. B. and Berkley, D. A.}, title = {Image method for efficiently simulating small-room acoustics}, journal = JASA, volume = {65}, pages = {943-950}, year = {1979} } and maybe (at)article{Neely79, author = {Neely, S. T. and Allen, J. B.}, title = {Invertability of a room impulse response}, journal = JASA, volume = {66}, pages = {165-169}, year = {1979} } -- Jont B. Allen, 973/360-8545 voice, 775/796-9844 (fax), 908/789-9575 (home fax) Technology Leader Speech Processing Software and Technology Research; Room E161, AT&T Labs-Research, Shannon Laboratory 180 Park Ave., Florham Park NJ, 07932-0971 ``It is hard to abandon the feeling that the unfamiliar is absurd and illogical.'' --G.A. Miller, p. 5 of his book `Language and communication'

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University