newbie here so please be kind if my question is stupid,

i'm making a patch for an audio installation,

I have 3 audio channels output by 3 speakers and

1 audio channel getting environment sound from a microphone placed is in the same area of the speakers.

The environmental soundlevel controls the reverb and volume of the audio that goes to the 3 speakers,

so it has some feedback/leakage, not real audio feedback, only "dynamic", because when the audio channels raise in volume the mic perceive a raise in the env_soundlevel.

I want to subtract the audio that go to the speakers from the audio that is received by the mic channel, so to have in input only (more or less) the "real" environmental soundlevel.

I looked in the tutorials and guides but found no solution.

Hi,
newbie here so please be kind if my question is stupid,
i'm making a patch for an audio installation,
I have 3 audio channels output by 3 speakers and
1 audio channel getting environment sound from a microphone placed is in the same area of the speakers.
The environmental soundlevel controls the reverb and volume of the audio that goes to the 3 speakers,
so it has some feedback/leakage, not real audio feedback, only "dynamic", because when the audio channels raise in volume the mic perceive a raise in the env_soundlevel.
I want to subtract the audio that go to the speakers from the audio that is received by the mic channel, so to have in input only (more or less) the "real" environmental soundlevel.
Is it technically possible?
I looked in the tutorials and guides but found no solution.
Forgive me if it's trivial,
i'm just learning,
thanks
gio


audio-subtraction

it's not really possible in praksis - as the signal comming into the microphone will be very different from the sound sent to the speakers....it can be done, as long as the signal is still in the digital domain, i.e. in the processing chain inside max...... if there is just 1 ms of delay between the to sounds being subtrated, a big part og the sound will pass thru...

so by trying this with a live input, you'll just get a weird phasing effect...

A solution would be to figure out where in the sound spectrum the feedback is a problem....possibly around 8000 hz and below 30... and use EQ and limiting...

> I want to subtract the audio that go to the speakers from the audio 

> that is received by the mic channel, so to have in input only (more 

> or less) the "real" environmental soundlevel. Is it technically 

Practically not, you get all kinds of filtered reflections which are not 

predictable, at least if you let people go into the space and have some 

What is the loudness of the speakers compared to the level of the other 

sounds? If the speakers are much softer, then it should not affect it 

too much, elsewise you would need to look only at frequency bands which 

If you create a negative feedback: the speakers get softer if the sound 

in the room gets louder, then you would also not run into problems.

> I looked in the tutorials and guides but found no solution.

> Forgive me if it's trivial, i'm just learning,

Its not trivial, so I can't forgive you unfortunately ;-)

KK wrote:
> I want to subtract the audio that go to the speakers from the audio 
> that is received by the mic channel, so to have in input only (more 
> or less) the "real" environmental soundlevel. Is it technically 
> possible?

Practically not, you get all kinds of filtered reflections which are not 
predictable, at least if you let people go into the space and have some 
air movement etc.

What is the loudness of the speakers compared to the level of the other 
sounds? If the speakers are much softer, then it should not affect it 
too much, elsewise you would need to look only at frequency bands which 
are not coverd by the speaker sound.
If you create a negative feedback: the speakers get softer if the sound 
in the room gets louder, then you would also not run into problems.

> I looked in the tutorials and guides but found no solution.
>  Forgive me if it's trivial, i'm just learning,

  [][]  [][][]  [][]  [][][]
[][][][][][][][][][][][][][][]

         Stefan Tiedje
         Klanggestalter
     Electronic Composition
               &
         Improvisation

            /~~~~~
     \   /|() ()|
     ))))) )|  |  |( \
     ///     _/)/ )))))
             ___/   ///

-------------------------x----
--_____-----------|-----------
--(_|_ ----|-----|-----()----
-- _|_)----|-----()-----------
----------()------------x-----

14, Av. Pr. Franklin Roosevelt,
94320 Thiais, France
Phone at CCMIX +33-1-57 42 91 09


Hi, thanks for the quick and precise reply,

>>> you get all kinds of filtered reflections which are not

predictable, at least if you let people go into the space and

+++ It is an outdoor installation so shouldn't be a lot of reverberations-reflections if I can place the mic in a good position, anyway the important thing that makes everything easier is that I don't need the audio, only the LEVEL, so it's not necessary to have the most precise subtraction, even an approximative value is good.

>>>What is the loudness of the speakers compared to the level of the other sounds? If the speakers are much softer, then it should not affect it too much, elsewise you would need to look only at frequency bands which are not coverd by the speaker sound.If you create a negative feedback: the speakers get softer if the sound in the room gets louder, then you would also not run into problems.

+++ Sometimes speaker sounds are softer than environment, sometimes much louder, creating a sort of "dynamic feedback" (when speaker sounds are louder -> mic gets a higher level-> making the speaker sounds louder and louder.) And I can't use negative feedback cause the function of the installation is exactly the opposite: when the environmental sound level raise, the music volume must raise.

I'll try out different solutions and let you know,

Hi, thanks for the quick and precise reply,
more details:
>>> you get all kinds of filtered reflections which are not
predictable, at least if you let people go into the space and
have someair movement etc.
+++ It is an outdoor installation so shouldn't be a lot of reverberations-reflections if I can place the mic in a good position, anyway the important thing that makes everything easier is that I don't need the audio, only the LEVEL, so it's not necessary to have the most precise subtraction, even an approximative value is good.

>>>What is the loudness of the speakers compared to the level of the other sounds? If the speakers are much softer, then it should not affect it too much, elsewise you would need to look only at frequency bands which are not coverd by the speaker sound.If you create a negative feedback: the speakers get softer if the sound in the room gets louder, then you would also not run into problems.
+++ Sometimes speaker sounds are softer than environment, sometimes much louder, creating a sort of "dynamic feedback" (when speaker sounds are louder -> mic gets a higher level-> making the speaker sounds louder and louder.) And I can't use negative feedback cause the function of the installation is exactly the opposite: when the environmental sound level raise, the music volume must raise.  

I'll try out different solutions and let you know,
thanks a lot for your precious support,
have y'all bright and peaceful days,
gio


audio subtraction