W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2004

audio contrast

From: David MacDonald <befree@magma.ca>
Date: Sat, 30 Oct 2004 23:23:48 -0400
Message-Id: <200410310323.i9V3NrW6009599@mail1.magma.ca>
To: <w3c-wai-gl@w3.org>, <wendy@w3.org>
HI All

 

I had an action item to create (good & bad) examples of audio contrast where
the foreground is 20db louder than the background and also where the
foreground is less than 20 db above the background.

 

I recorded these tracks and set the various levels by measuring the RMS
(root mean squared) power for the foreground and background of each track in
each example. Then mixed them together in wave format and burned them to
MP3. The RMS is the average volume level. In the first example the voice
(foreground) is recorded at -17.52 decibels and the music (background) is at
-37.52 decibels, which makes the foreground 20 decibels louder than the
background. In the second example the voice (foreground) is at -18 decibels
and the music (background) is at about -16 decibels making the foreground
only 2 decibels louder than the background.

 

The following is the link to the examples

 

http://www.eramp.com/david/audio_contrast.htm 

 

I had a meeting with the forensic audio engineer for the Royal Canadian
Mounted Police (RCMP) and discussed with him the results of my examples. Our
WCAG 2.0 document says if the foreground is 20db above the background then
it is about 4x the volume of the background. 20db is a huge drop in volume,
and as can be heard on my example. It will be hard to get people who produce
sound tracks to buy into our guidelines given this ratio, I think. Movies,
and commercials etc. will all fail currently.  I have a hard of hearing
friend who hates background noise and music and would like to see it
eliminated. Our 20db recommendation just about does that. But it is at level
3 so perhaps it is OK to leave it. I think though that we could do a little
more consultation with the hard of hearing community explaining that the
rate of adoption of our guideline may be proportional to our level of
requirement for content providers to place the foreground above the
background. Or maybe in the future someone can make a standard that says the
foreground and background tracks are delivered separately to media players
and the user can control the audio contrast r on each media device the way
they currently control bass and treble. Sort of like the requirement for
line 21 captioning. Hmmm any policy makers out there?

 

The ABC's of Decibels: 

 

Power, volume and perceived volume are not the same.  

 

Perception of loudness is logarithmic. All things being equal, when you
double the amount of power that you feed to your speakers, you get a
3-decibel increase in the sound pressure level at the output: 3 dB is
considered to be a noticeable difference in volume. A 10-dB increase is
twice as loud. Therefore, the difference between 1 and 2 watts is equal to
the difference between 10 and 20 watts-likewise for 100 and 200 watts. 

 

"The human body perceives ten times the sound-power as being only twice as
loud. And 100 times the sound power is only 4 times as loud to our ears.
Weird but true. And actually good for us, probably designed to that effect
by evolution, so that we can hear and effectively use the extremely wide
range of sound 
powers that frequently exist in various times and places."
http://www.newton.dep.anl.gov/askasci/eng99/eng99325.htm 
 

The formula for the difference in decibels between two different sound
sources is defined to be 

 

10 log (P2/P1) dB            (where the log is to base 10)

 

If the foreground was 4x louder (perceived volume) than the background, it
would need to be 100 times the power of the background. 

 

The formula would be

10 (log (100/1)) db 

= 10 *(log 100) db

= 10 * 2db

= 20db

 

 

Cheers

David MacDonald

------------------------------------------

Access empowers people...
         .Barriers disable them.

 

www.eramp.com

 

 
Received on Sunday, 31 October 2004 03:23:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:32 GMT