[PD] 3d sound with 2 speaker

matthew jones mj at isvr.soton.ac.uk
Thu Jan 6 14:07:09 CET 2005


oh dear, to get anything realistic at all would be extremely involved.
There is a system here (University of Southampton) that people have 
developed over years that uses two loudspeakers placed VERY close to each 
other and works well for a few people sitting in the sweet spot.
Thing is it uses a supercomputer to do realtime convolution of many 
individual sounds with very long FIR filters...

The method is basically three stage:
-use a kemar (dummy head) and the exact loudspeaker setup to measure the 
transfer functions from loudspeaker to the listeners ears
-use a kemar to measure the head-related transfer functions (HRTF's) for a 
whole bunch of directions of arrival
-use these frequency responses to (a) invert the louspeaker->ear response 
and (b) mimic the HRTF response, so that sounds when they finally arrive at 
your ear are as though they coming from the desired direction.

The reason the speakers are placed so close together is that it allows you 
to more easily control the sound an INDIVIDUAL ear hears, ie. it cancels the 
crosstalk for a larger region of space around the ear.
Just some info to demonstrate the HUGENESS of the problem!
good luck,

Matt

-=-=-=-=-=-=-=-=-=-=-=-
http://www.loopit.org
-=-=-=-=-=-=-=-=-=-=-=-
----- Original Message ----- 
From: "pun chik" <punchik at fastmail.fm>
To: <pd-list at iem.at>
Sent: Wednesday, January 05, 2005 9:27 PM
Subject: [PD] 3d sound with 2 speaker


>
>
> hi, do anybody have tried creating doppler effects with pd? i was
> wondering if there is any external or abstraction, etc to  create a
> surrounding effect within a stereo field. id like to create a patch
> using 2 speakers where a sound (an insect) flyes around you. any
> hint????
>
>
> bye
>
> punchik
> -- 
>  pun chik
>  punchik at fastmail.fm
>
>
> _______________________________________________
> PD-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> 
> http://iem.at/cgi-bin/mailman/listinfo/pd-list 





More information about the Pd-list mailing list