Regarding Spacious Stereo Width..

Apr 9, 2008 at 9:11am

Regarding Spacious Stereo Width..

Ive been trying to figure out for a while now, different ways of making my reverb and other kinds of sounds a bit more wider in stereo, making the reverb more spacious etc.

As ive been hearing a lot of atmos/drones made out of max/msp, with just this spacial stereo widening (for example the ever so great: http://music.columbia.edu/~brad/mm-2/) i thought id pop the question out there, Do you have any special tricks for doing this? Or perhaps you could hand me some hints on the matter?

Is it all just filtered and pan’ed delays, and some feedback goin on?

#36808
Apr 9, 2008 at 9:42am

Hi,

One trick is decorrelation, check out

# The Decorrelation of Audio Signals and Its Impact on Spatial Imagery
# Gary S. Kendall
# Computer Music Journal, Vol. 19, No. 4 (Winter, 1995), pp. 71-87


Owen

steffensen wrote:
> Ive been trying to figure out for a while now, different ways of
> making my reverb and other kinds of sounds a bit more wider in
> stereo, making the reverb more spacious etc.
>
> As ive been hearing a lot of atmos/drones made out of max/msp, with
> just this spacial stereo widening (for example the ever so great:
> http://music.columbia.edu/~brad/mm-2/) i thought id pop the question
> out there, Do you have any special tricks for doing this? Or perhaps
> you could hand me some hints on the matter?
>
> Is it all just filtered and pan’ed delays, and some feedback goin on?
>
>

#126518
Apr 9, 2008 at 9:54am

#126519
Apr 9, 2008 at 10:04am

Even tho i dont have that book at hand, i found some snippets from it on google, and i can see the subject it presents. Its almost some kind of chorusing if i understand it right.

Most interesting.. i think im gonna try panning the signal out to different random-pitchmodifying effects, and only slightly affecting the different signals.

Good idea, and a good tip, thank you.

#126520
Apr 9, 2008 at 10:10am

jmdarremont,
Looks interesting! To bad i dont speak code, not yet anyway. :)

#126521
Apr 9, 2008 at 10:31am

I guess you’re replying to me?

It’s a journal article; if you’re not affiliated with a university and
can’t find it on-line, you should be able to order a copy via your local
library for a small charge, if they’re anything like the ones here in
the UK.

It’s not a chorusing effect – in fact, when it works well, the perceived
spectral qualities of the input are minimally effected.

How it works is by the passing the input through two all-pass filters
(i.e. that only mess with phase) whose coefficients have been designed
to give phase responses that are random and minimally correlated, thus
reducing the correlation of the input phase between channels and
increasing perceived spatial width. It can also be used for reducing
the spectral effects of short delays.

A plus is that mono sources can be spatialized, which doesn’t work with
Mid-Side processing (as there’s no side signal). A drawback is that the
perceived point of origin of the sound is smeared. Also, the same sets
of filter coefficients don’t work equally well for all inputs, so a
certain amount of trial and error is required.


Owen

steffensen wrote:
> Even tho i dont have that book at hand, i found some snippets from it
> on google, and i can see the subject it presents. Its almost some
> kind of chorusing if i understand it right.
>
> Most interesting.. i think im gonna try panning the signal out to
> different random-pitchmodifying effects, and only slightly affecting
> the different signals.
>
> Good idea, and a good tip, thank you.
>
>

#126522
Apr 9, 2008 at 10:51am

Yes, the reply was ment for you, sorry.

Aha, i see, phase trix. Ill see what i can make out of that. Just the thought of this makes total sense!

Thanks!

Quote: owen wrote on Wed, 09 April 2008 04:31
—————————————————-
> I guess you’re replying to me?
>
> It’s a journal article; if you’re not affiliated with a university and
> can’t find it on-line, you should be able to order a copy via your local
> library for a small charge, if they’re anything like the ones here in
> the UK.
>
> It’s not a chorusing effect – in fact, when it works well, the perceived
> spectral qualities of the input are minimally effected.
>
> How it works is by the passing the input through two all-pass filters
> (i.e. that only mess with phase) whose coefficients have been designed
> to give phase responses that are random and minimally correlated, thus
> reducing the correlation of the input phase between channels and
> increasing perceived spatial width. It can also be used for reducing
> the spectral effects of short delays.
>
> A plus is that mono sources can be spatialized, which doesn’t work with
> Mid-Side processing (as there’s no side signal). A drawback is that the
> perceived point of origin of the sound is smeared. Also, the same sets
> of filter coefficients don’t work equally well for all inputs, so a
> certain amount of trial and error is required.
>
> —
> Owen
>
>
>
> steffensen wrote:
> > Even tho i dont have that book at hand, i found some snippets from it
> > on google, and i can see the subject it presents. Its almost some
> > kind of chorusing if i understand it right.
> >
> > Most interesting.. i think im gonna try panning the signal out to
> > different random-pitchmodifying effects, and only slightly affecting
> > the different signals.
> >
> > Good idea, and a good tip, thank you.
> >
> >
>
>
>
—————————————————-

#126523
Apr 9, 2008 at 1:11pm

Wow, thanks for the nice words!

I was (still am) a major fan of Arif Mardin’s production techniques,
and one of the tricks he used was a short single delay, original and
delayed signal panned hard L/R, before sending the sound into a good
reverb. If the delay is really short it does a cute ‘precedence
effect’ thing and makes it sound like the source is way off to the
left or right of the stereo field, and if it is a little bit longer
it makes it seem like you have a really big head. :-) No feedback,
btw.

brad

http://music.columbia.edu/~brad

On Apr 9, 2008, at 5:11 AM, steffensen wrote:

>
> Ive been trying to figure out for a while now, different ways of
> making my reverb and other kinds of sounds a bit more wider in
> stereo, making the reverb more spacious etc.
>
> As ive been hearing a lot of atmos/drones made out of max/msp, with
> just this spacial stereo widening (for example the ever so great:
> http://music.columbia.edu/~brad/mm-2/) i thought id pop the
> question out there, Do you have any special tricks for doing this?
> Or perhaps you could hand me some hints on the matter?
>
> Is it all just filtered and pan’ed delays, and some feedback goin on?

#126524
Apr 9, 2008 at 7:00pm

Thanks for noticing!
Nice to see you around. :)

Good tip there, that was kind of what i ment before by panned and filtered delays.
I really do get good results by doin this on the actual sound. Using this technique on a reverb makes it kind of metallic however, even when filtering the different delayed signals.

Anyway, ill find a way im sure.
Thanks again!

Quote: Bradford Garton wrote on Wed, 09 April 2008 07:11
—————————————————-
> Wow, thanks for the nice words!
>
> I was (still am) a major fan of Arif Mardin’s production techniques,
> and one of the tricks he used was a short single delay, original and
> delayed signal panned hard L/R, before sending the sound into a good
> reverb. If the delay is really short it does a cute ‘precedence
> effect’ thing and makes it sound like the source is way off to the
> left or right of the stereo field, and if it is a little bit longer
> it makes it seem like you have a really big head. :-) No feedback,
> btw.
>
> brad
> http://music.columbia.edu/~brad
>
>
> On Apr 9, 2008, at 5:11 AM, steffensen wrote:
>
> >
> > Ive been trying to figure out for a while now, different ways of
> > making my reverb and other kinds of sounds a bit more wider in
> > stereo, making the reverb more spacious etc.
> >
> > As ive been hearing a lot of atmos/drones made out of max/msp, with
> > just this spacial stereo widening (for example the ever so great:
> > http://music.columbia.edu/~brad/mm-2/) i thought id pop the
> > question out there, Do you have any special tricks for doing this?
> > Or perhaps you could hand me some hints on the matter?
> >
> > Is it all just filtered and pan’ed delays, and some feedback goin on?
>
>
—————————————————-

#126525

You must be logged in to reply to this topic.