Subject: Re: Multi-speaker ...
From: Alexandra Hettergott (a.hettergott@wanadoo.fr)
Date: Wed Mar 21 2001 - 12:55:07 EST
Kevin Austin replying to Lindsay Manning :
>>I was always a little skeptical about Bregman's suggestions on this
>>because I don't think he gives localization enough credit.
>Hmmmm ... a somewhat related topic has appeared recently on AUDITORY.
>One of the points that Bregman brings out is that most people have two
>eardrums and that we do all of this work of segregation, fusion,
>streaming etc etc based upon _two_ simple displacements.
Well, Imho that localization clues are taken into consideration if being
(obtainable and) sufficiently pertinent in relation to a specific
(situational) information ; just with regard to segregation (and with
visual / gestual help) this might apply to both the cocktail party /
orchestra situation a single voice being recognized / segregated from a
polyphonic context *easier* by means of just this (visually-aided)
localization clue. The argument that here it is not audition *only* in a
certain sense I think is neglect-able in view of the fact that this
topic is usually being discussed *just* with respect to "survival in
wilderness" (hence a claim to (situative) "reality") -- and the visual
sense -- as Kevin has already been pointing out -- seems to really
represent the better "localizer" in wo/man ...
As for ea being reproduced by non-moving objects like loud-speakers,
giving parallel (source/channel-conform) visual clues thus should really
be helpful in this regard ...
:ah
This archive was generated by hypermail 2b27 : Wed Jun 11 2003 - 13:06:57 EDT