Audio McGurking
Arny Krueger
One of the more striking audible/visual illusions is the
McGurk
effect. The McGurk effect shows that hearing is not
believing, when
there is also conflicting visual evidence. In fact, visual
evidence
can completely overcome things that are otherwise clearly
audible when
presented without the distracting visual evidence.
You can even be fully aware of the McGurk effect and its
application
to your immediate situation, and still be fooled.
The "McGurk effect" was first described by Harry McGurk and
John
MacDonald in "Hearing lips and seeing voices", Nature 264,
746-748
(1976). 1976? This is very old news!
An audio recording is played of a person saying a certain
thing. A
synchronized video of that person saying something else is
displayed.
Almost all of the time, almost all people perceive a sound
that
seems to match the video. Surprisingly what they perceive
isn't the sound that
is there for them to hear.
If you listen to the sound track with your eyes shut, you
can hear
the audio recording quite accurately. Open your eyes again,
and you
perceive a sound that matches the video. What you perceive
does not
match the audio.
It's amazing that when even when you know the trick and
exactly how
it applies to the current situation, you can still be fooled
again
and again. This has happened to me many times. This is not
delusion,
it is illusion.
One of the best web-based demos of the McGurk effect I've
seen can
be found at
http://www.media.uio.no/personer/arntm/McGurk_english.html .
Since many common English words that sound alike can mean
different
things, the McGurk effect can have some striking but highly
confusing effects.
McGurk's effect is so strong that sounds don't have to be
hidden
away in words for it to confuse things. As the web
demonstration
shows, even isolated syllables can be strongly impacted by
the
McGurk effect.
Imagine a comedy sketch based on the McGurk effect. Two
people have
a humorous conversation based on they words they actually
say, but
the audience sees visuals of the performers saying something
else.
This little trick could be quite shocking if strong
profanity or
other highly inflammatory statements were put into the
audience's
perceptions by means of simple visual effects. Seems like a
natural
for Letterman or SNL.
The sound track would be clear evidence that the performers
said
nothing wrong, but the telephones at the FCC and network
headquarters would no doubt light up like the Detroit River
on
fireworks night! It might be interesting to have a legal
test of
comedic McGurking.
Applications of the McGurk effect to sighted evaluations of
audio
components seem quite clear. During most sales presentations
and
home demonstrations, listeners are given visual information
indicating that sound quality has changed, usually that
sound
quality is greatly improved.
The visual information that is presented during audio
equipment
demonstrations is often quite elaborate. Consider a
comparison of a
vacuum-tube power amp with its richly glowing vacuum
bottles, and a
solid state power amp in a darkened plain metal box.
Consider a high
end vinyl playback system with artistic polished metal
shapes and a
deep ebony disc spinning hypnotically, as compared to a
dinky little
digital player with a tiny ugly light green glowing screen
on its
front panel.
We know from our level-matched, time-synched, blind
listening
comparisons that the audible cues are often subtle, to say
the
least. Therefore it is no surprise that visual evidence can
lead to
perceptions that differ from the sound that is in the
listening room.
Considering what we know about the McGurk effect, it is easy
to
understand why people will report perceptions that agree
with the
visual cues that they receive no matter what sound is in the
room.
Knowing about the McGurk effect helps me understand why so
many
people were fooled by the SET amp and vinyl demonstrations I
saw at
HE2005 in New York a few months ago.
McGurk's effect shows that visual information is far
stronger than
would merely suffice to cause people to perceive that one
amplifier
sounds different from another when they actually sound
similar or
alike. McGurk's effect is capable of making people believe
they hear
something that is quite different, even something that is
almost the opposite from the
sound that is actually in the room.
Audio McGurking might explain perceiving favorable sound
quality from
a gritty-sounding vacuum tube power amp, even though the SS
power
amp is sonically superior. Or, Audio McGurking might explain
why so
many perceive that vinyl sounds better than good digital.
Sighted evaluations are something like the fundamental
principle of
stage magic which is distracting the audience away from
what's
really happening, towards what the performer wants the
audience to
perceive is happening.
Jem Raid
of dosh now, mine cost a fiver.
I thought it sounded OK until I bought an SS amp, can't remember the name.
I then realised that the Rogers was as soft as shit.
Jem
Dave Plowman (News)
Any chance of setting your wordwrap to accepted Usenet standards?
--
*Remember, no-one is listening until you fart.*
Dave Plowman da...@davenoise.co.uk London SW
To e-mail, change noise into sound.
Dave Plowman (News)
Nothing wrong with them if you have efficient speakers and keep it within
its rated output. In original condition it will sound really very nice -
apart from the rather noisy pre-amp stages by modern standards.
However most would find the power output *very* limited for today.
--
*If you're not part of the solution, you're part of the precipitate *
Arny Krueger
message news:4d8ead9...@davenoise.co.uk
> Arny,
> Any chance of setting your wordwrap to accepted
> Usenet standards?
The problem was that I had OE and OE-Quotefix set to
different wrap points.
Rob
> Audio McGurking
>
<snip more because of peculiar line breaks than content>
>
> Audio McGurking might explain perceiving favorable sound
> quality from
> a gritty-sounding vacuum tube power amp, even though the SS
> power
> amp is sonically superior. Or, Audio McGurking might explain
> why so
> many perceive that vinyl sounds better than good digital.
>
> Sighted evaluations are something like the fundamental
> principle of
> stage magic which is distracting the audience away from
> what's
> really happening, towards what the performer wants the
> audience to
> perceive is happening.
>
Cuts both ways I suppose - some* people just might have a loaded
impression that SS 'AA Mosfet super linear balanced open loop zero
feedback' might intinsically sound better, particularly if they're left
with a brochure in their hand or it's emblazoned on the shiny front
panel of the amplifier. Similarly CD and DAB - weren't we told something
along the lines of 'perfect sound forever'?
Environmental cues are bound to influence perception - whoever you are
and whatever you listen through.
b/w
Rob
* partially sighted and blind people would provide an interesting
perspective here.
Jem Raid
>
> Rob
>
> * partially sighted and blind people would provide an interesting
> perspective here.
>
>
Years ago I used to raise money for tandems for the blind, one of the
organisers also belonged to the Ramblers Association, and she used to make
recordings of the walks she went on, she played part of one for me once, I
was astounded, it was like being there.
I've a mind to buy a tape recorder and have a go myself.
Pardon the off topic, but the memory of it is very strong.
Jem
Ethan Winer
I agree with all of that, and here's an alternate thought on one item:
> Audio McGurking might explain why so many perceive that vinyl sounds
better than good digital. <
When people prefer the sound of vinyl (and analog tape too, especially if
you hit it hard) I think it may be some aspect of the distortion that's
appealing. I've taken mixes from my DAW and recorded them to a cassette and
noticed an "improvement" of sorts. Obviously this is an effect, and not
higher fidelity. But for some reason the music can sound more coherent, for
lack of a better word. This works best with sparse music. With dense busy
mixes the added distortion tends to muddy things up further, especially the
IM components.
Or maybe people grew up listening to music full of scratches and pops, and
when those are missing the music somehow seems lacking?
--Ethan
Harry Lavo
"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
news:rtqdne6RXO_...@giganews.com...
No, it's 2nd harmonic distortion, which in small doses, leads to enhanced
presence in the midrange. Used along with phasing in the "vocal enhancers"
used in studios since the '70's.
You can get a similar effect on phonograph by lowering the VTA just below
its optimum point. The sound will "dull out" but the voices and instruments
will have more body.
Arny Krueger
> I agree with all of that, and here's an alternate thought
> on one item:
>> Audio McGurking might explain why so many perceive that
>> vinyl sounds better than good digital. <
> When people prefer the sound of vinyl (and analog tape
> too, especially if you hit it hard) I think it may be
> some aspect of the distortion that's appealing.
To me, the colorations of vinyl and analog tape are vastly
different. So much so that in the days of, they made good
complements to each other. Analog tape shaved off the
dynamics, making a lot of wide dynamic range *safe* for
vinyl and the limited dynamic playback gear of the day.
> I've taken mixes from my DAW and recorded them to a
cassette
> and noticed an "improvement" of sorts. Obviously this is
> an effect, and not higher fidelity. But for some reason
> the music can sound more coherent, for lack of a better
> word. This works best with sparse music. With dense busy
> mixes the added distortion tends to muddy things up
> further, especially the IM components.
Most likely cause of music sounding tighter and more
coherent after a trip through a cassette seems to me to be
the compression that is build into analog tape.
> Or maybe people grew up listening to music full of
> scratches and pops, and when those are missing the music
> somehow seems lacking?
I've heard enough vinyl with the tics and pops digitally
removed quite effectively, to doubt that.
Ashley
http://www.fleximusic.com/waveditor/audioeditor.htm
"Arny Krueger" <ar...@hotpop.com> wrote in message
news:9M-dnVEHaJk...@comcast.com...