In article , Don Pearce
wrote:
On Wed, 23 Jun 2004 10:14:07 +0100, Jim Lesurf
wrote:


Bear in mind that what I've written is very much a *simplified* version
of what the physiologists now use as a model, but even so it has some
surprising implications for audio/hifi... :-)

Jim, I remember well a demo I heard many years ago. The sound was of a
person reading words out slowly against a noise background. The
background was sufficiently loud that it was impossible to work out the
words. We listened several times, but still no intelligibility The
proctor then told us what a couple of the words actually were. As if by
magic, these words were suddenly clearly audible against the noise,
whilst the rest remained inaudible.

Not come across the above, but it sounds like a neat example of how humans
can perform 'pattern recognition'. I'd presume this was occuring at a stage
later than the cochlea, though. Somewhere in the brain activity. Afraid
that I don't really know much about that beyond noting that humans seem to
have some remarkable abilities in such things.

I suppose that it is analogous in some ways to the visual test where a
sheet of white paper with blobs on it can suddenly be 'recognised' as a
dalmation dog against a background of leaves. This presumably only works if
you are familiar with what a dog/dalmation looks like! :-)

I don't suppose you know where a copy of this test may be found? It is
really impressive.

Afraid not. Even my own brain activity baffles me. :-)

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

On Wed, 23 Jun 2004 11:56:53 +0100, Jim Lesurf
wrote:

In article , Don Pearce
wrote:
On Wed, 23 Jun 2004 10:14:07 +0100, Jim Lesurf
wrote:


Bear in mind that what I've written is very much a *simplified* version
of what the physiologists now use as a model, but even so it has some
surprising implications for audio/hifi... :-)

Jim, I remember well a demo I heard many years ago. The sound was of a
person reading words out slowly against a noise background. The
background was sufficiently loud that it was impossible to work out the
words. We listened several times, but still no intelligibility The
proctor then told us what a couple of the words actually were. As if by
magic, these words were suddenly clearly audible against the noise,
whilst the rest remained inaudible.

Not come across the above, but it sounds like a neat example of how humans
can perform 'pattern recognition'. I'd presume this was occuring at a stage
later than the cochlea, though. Somewhere in the brain activity. Afraid
that I don't really know much about that beyond noting that humans seem to
have some remarkable abilities in such things.

I suppose that it is analogous in some ways to the visual test where a
sheet of white paper with blobs on it can suddenly be 'recognised' as a
dalmation dog against a background of leaves. This presumably only works if
you are familiar with what a dog/dalmation looks like! :-)

I don't suppose you know where a copy of this test may be found? It is
really impressive.

Afraid not. Even my own brain activity baffles me. :-)

Slainte,

Jim

I think there is an analogy to telecoms - namely CDMA. What you
receive is noise, unless you know what the spreading code is. Then you
can perform what looks rather like magic on the signal to noise ratio.
Nonsense, of course, since what actually happens is that the
information hasn't been transmitted - it is already present at both
ends of the link. Impressive though, until you suss what is going on.

d
Pearce Consulting
http://www.pearce.uk.com

Arny Krueger wrote:

More than a microphone, the ear embodies a pretty interesting spectrum
analyzer as well.

....and if you take enough drugs you get to see the graphical display that
goes with the analyser :-))



--
The presence of this sig indicates that I'm under the influence of
excess alcohol. Until I'm sober enough to remember to switch this sig
file off, please treat the above as merely drunken ranting and treat it
with the contempt it probably deserves.

I apologise in advance for any offence caused :-)