Re: streams and groups (lonce wyse )


Subject: Re: streams and groups
From:    lonce wyse  <lwyse(at)MINDMAKER.COM>
Date:    Thu, 17 May 2001 02:31:01 +0800

Venerable List, Regarding clapping & applause: My experience with clappers and applause comes from building sound models for use in interactive applications. Just to make sure this point isn't missed (as it hasn't been mentioned yet), the periodicity of the signal from individual clappers - even when the period is slightly different across clappers) helps the listener keep track of the different sources. If the timing is perturbed away from periodicity while keeping the overall rates the same, the perception of the aggregate changes dramatically. An interesting question is: at what number of clappers does the periodicity of the individual not matter anymore? Individual clap sounds, even those from a single clapper, are quite different in frequency content. So, another interesting question comes up in modeling: How many different single-clap sounds does one need to create applause? (A low number is quite a-plausible!) If I remember correctly, we needed about 30 in dense applause in order not to hear the "buzz" of precisely repeating samples. - lonce


This message came from the mail archive
http://www.auditory.org/postings/2001/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University