aonghusdav@yahoo.co.uk

any idea on how max/msp could be used to find a persons location in a room using four microphones?

location

You could use a fiddle~ external on each of these microphones which will give you a number value for velocity. You could simply use an if statement to choose the one that was loudest, else if you were feeling adventurous you could come up with some matrix system where you could subdivide it between the values.

Might be easier to just use jitter and cyclops or something maybe? but depends waht you're trying to do- and with cyclops involved, budget!

You could use a fiddle~ external on each of these microphones which will give you a number value for velocity. You could simply use an if statement to choose the one that was loudest, else if you were feeling adventurous you could come up with some matrix system where you could subdivide it between the values.
Might be easier to just use jitter and cyclops or something maybe? but depends waht you're trying to do- and with cyclops involved, budget!
T


> any idea on how max/msp could be used to find a persons location in a room using four microphones?

I don't beleive it s a reliable way of tracking a person in a room.

the way the sounds reflect on the walls, their directions of emition will interfeer to much in the

I suggest you go for a motion tracking solution using a camera. if you want the room to be dark

It'll be easier, clearer and a lot more precise and efficient.

I don't beleive it s a reliable way of tracking a person in a room.
the way the sounds reflect on the walls, their directions of emition will interfeer to much in the
analysis you want to do.
I suggest you go for a motion tracking solution using a camera. if you want the room to be dark
use infra-red.

You could also try introducing a continuous, high frequency sound into the room. Then watch for level changes from the microphone's input as the person casts a "shadow" moving around the room. High freq's are directional, so you might need to point a speaker at each microphone. I've used this technique in a studio setting before with one speaker and mic. I also recall seeing a Nord Modular patch that used the technique. You might try searching a Nord forum.

> any idea on how max/msp could be used to find a persons location in a room

difficult as it would require the person to make significant noise,

preferably percussive noise (then you could try to track delays).

It would also only work with sensing noise levels, if the room is as dry

as possible, as the reverberation would level out the differences in

volume caused by the distance to the microphones...

Some skills in basic acoustics are required...

I don't know if anybody built a ultrasonic sensor around a normal

soundcard (at 96 or 192 kHz) It would require special

aongpaint wrote:
> any idea on how max/msp could be used to find a persons location in a room

difficult as it would require the person to make significant noise,
preferably percussive noise (then you could try to track delays).
It would also only work with sensing noise levels, if the room is as dry
as possible, as the reverberation  would level out the differences in
volume caused by the distance to the microphones...

I don't know if anybody built a ultrasonic sensor around a normal
soundcard (at 96 or 192 kHz) It would require special
speakers/microphones though....

-- 
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com