| Author |
Message |
Joe Kesselman
Guest
|
 Posted:
Tue Nov 01, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
| Quote: | Did you try dithering and dithering and dithering... and dithering and
dithering?
|
Oh, I'm _good_ at dithering. That's why I haven't invested in more
hardware in quite some time...
|
|
| Back to top |
|
 |
SSJVCmag
Guest
|
| Posted:
Tue Nov 01, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
On 10/31/05 5:29 PM, in article OUw9f.63756$y_1.27732@edtnps89, "Lorin David
Schultz" <Lorin@DAMNSPAM!v5v.ca> wrote:
| Quote: | "Ty Ford" <tyreeford@comcast.net> wrote:
Can you dither that?
Ty Ford
Speaking of that...
Sometimes while doing sound design work I come across a sound or effect
that would work better if certain elements were less "distinct." For
example, I recently had to work with a whooshy kind of pass-by sound
that was perfect except for a kind of grainy character that couldn't be
removed with EQ. While trying a fast decaying reverb, I joked, "I wish
Pro Tools had the same blur feature as Photoshop!"
Now I'm wondering if there is an equivalent... ideas?
|
I think maybe a place to start is the way a vocoder works (and of course
thus the various mp3 transforms) with varying numbers of bands and dynamics
resolutions and then blending that back in with the original...
And of course there's always KYMA... |
|
| Back to top |
|
|
hank alrich
Guest
|
| Posted:
Tue Nov 01, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
Lorin David Schultz wrote:
| Quote: | Sometimes while doing sound design work I come across a sound or effect
that would work better if certain elements were less "distinct." For
example, I recently had to work with a whooshy kind of pass-by sound
that was perfect except for a kind of grainy character that couldn't be
removed with EQ. While trying a fast decaying reverb, I joked, "I wish
Pro Tools had the same blur feature as Photoshop!"
Now I'm wondering if there is an equivalent... ideas?
|
Did you try dithering and dithering and dithering... and dithering and
dithering?
--
ha |
|
| Back to top |
|
|
Evangelos Himonides
Guest
|
| Posted:
Wed Nov 02, 2005 2:02 am Post subject:
Re: Gaussian blur for audio? |
|
|
I'm with Arny's suggestion and would add a little bit of chorus just in
order to introduce a somehow 'y axis' blur as well...
I cool piece of software called 'Coagula' can be found here:
http://www.sonicspot.com/coagula/coagula.html
It's free... have a go!
Regards,
Evangelos
%
Evangelos Himonides
IoE, University of London
tel: +44 2076126599
fax: +44 2076126741
"Allas to those who never sing but die with all their music in them..."
Oliver Wendell Holmes
% |
|
| Back to top |
|
|
Jona Vark
Guest
|
| Posted:
Wed Nov 02, 2005 2:47 am Post subject:
Re: Gaussian blur for audio? |
|
|
"Lorin David Schultz" <Lorin@DAMNSPAM!v5v.ca> wrote in message
news:OUw9f.63756$y_1.27732@edtnps89...
| Quote: | "Ty Ford" <tyreeford@comcast.net> wrote:
Can you dither that?
Ty Ford
Speaking of that...
Sometimes while doing sound design work I come across a sound or effect
that would work better if certain elements were less "distinct." For
example, I recently had to work with a whooshy kind of pass-by sound
that was perfect except for a kind of grainy character that couldn't be
removed with EQ. While trying a fast decaying reverb, I joked, "I wish
Pro Tools had the same blur feature as Photoshop!"
Now I'm wondering if there is an equivalent... ideas?
|
FIR Filter.
| Quote: |
--
"It CAN'T be too loud... some of the red lights aren't even on yet!"
- Lorin David Schultz
in the control room
making even bad news sound good
(Remove spamblock to reply)
|
|
|
| Back to top |
|
|
Bob Cain
Guest
|
| Posted:
Wed Nov 02, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
Jona Vark wrote:
| Quote: | "Lorin David Schultz" <Lorin@DAMNSPAM!v5v.ca> wrote in message
news:OUw9f.63756$y_1.27732@edtnps89...
"Ty Ford" <tyreeford@comcast.net> wrote:
Can you dither that?
Ty Ford
Speaking of that...
Sometimes while doing sound design work I come across a sound or effect
that would work better if certain elements were less "distinct." For
example, I recently had to work with a whooshy kind of pass-by sound
that was perfect except for a kind of grainy character that couldn't be
removed with EQ. While trying a fast decaying reverb, I joked, "I wish
Pro Tools had the same blur feature as Photoshop!"
Now I'm wondering if there is an equivalent... ideas?
FIR Filter.
|
An FIR filter that "blurs" a sample by a set of nearby samples, such as
a Gaussian, typically effects an attenuation of high frequencies.
Bob
--
"Things should be described as simply as possible, but no simpler."
A. Einstein |
|
| Back to top |
|
|
Les Cargill
Guest
|
| Posted:
Wed Nov 02, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
Lorin David Schultz wrote:
| Quote: | "Ty Ford" <tyreeford@comcast.net> wrote:
Can you dither that?
Ty Ford
Speaking of that...
Sometimes while doing sound design work I come across a sound or effect
that would work better if certain elements were less "distinct." For
example, I recently had to work with a whooshy kind of pass-by sound
that was perfect except for a kind of grainy character that couldn't be
removed with EQ. While trying a fast decaying reverb, I joked, "I wish
Pro Tools had the same blur feature as Photoshop!"
Now I'm wondering if there is an equivalent... ideas?
|
Analog tape, at significant levels? ( I know, PITA... )
--
Les Cargill |
|
| Back to top |
|
|
Chris Hornbeck
Guest
|
| Posted:
Wed Nov 02, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
On Mon, 31 Oct 2005 22:29:02 GMT, "Lorin David Schultz"
<Lorin@DAMNSPAM!v5v.ca> wrote:
| Quote: | Sometimes while doing sound design work I come across a sound or effect
that would work better if certain elements were less "distinct." For
example, I recently had to work with a whooshy kind of pass-by sound
that was perfect except for a kind of grainy character that couldn't be
removed with EQ. While trying a fast decaying reverb, I joked, "I wish
Pro Tools had the same blur feature as Photoshop!"
Now I'm wondering if there is an equivalent... ideas?
|
It strikes me as interesting that our visual processing,
which occupies so much of our attention and which we
*believe* so completely, seems to enjoy being teased.
The grainier and the more monochromatic the more evocative;
like that.
Large portions of this must be cultural, of course,
and the audio analogs carry the same baggage, but
still.... Great question.
And I suspect that fractals will be at the core of any
answer.
Thanks, as always,
Chris Hornbeck |
|
| Back to top |
|
|
Bob Cain
Guest
|
| Posted:
Thu Nov 03, 2005 2:03 am Post subject:
Re: Gaussian blur for audio? |
|
|
Arny Krueger wrote:
| Quote: | In audio, this would be approximated by weighted samples
taken near the sample being blurred. The displacement would
be in time, not space.
This is basically reverb or chorus.
|
No, a weighted average of nearby samples suppresses high frequencies in
ways determined by the weights, in general a lowpass. To get reverb and
chorus you add in weighted samples from a considerable distance in the past.
Bob
--
"Things should be described as simply as possible, but no simpler."
A. Einstein |
|
| Back to top |
|
|
Bob Cain
Guest
|
| Posted:
Thu Nov 03, 2005 2:06 am Post subject:
Re: Gaussian blur for audio? |
|
|
SSJVCmag wrote:
| Quote: | An FIR filter that "blurs" a sample by a set of nearby samples, such as
a Gaussian, typically effects an attenuation of high frequencies.
And that;s analogous to the loss of edges and definitioon in a photographic
blur...
|
Right, I was trying to relate the audio effect of that to something we
are already familiar with, a lowpass.
Bob
--
"Things should be described as simply as possible, but no simpler."
A. Einstein |
|
| Back to top |
|
|
Ron Capik
Guest
|
| Posted:
Thu Nov 03, 2005 4:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
SSJVCmag wrote:
| Quote: |
...snip...
Right.
Mr Hornsbeck nailed this earlier there.. It's "defined variable local pixel
averaging" in specific implementations to achieve the result.
Not time-delay in any of it's usual forms
(echo/flange/chorus/reflection/reverb)
Again, using a vocoder (which is a band-defined-averaging system) is the
closest that we have and the implementation of this sort of audio coding as
seen (ok heard) in mp3's is a good HANDY place to start... I'd shove the
things through various medium and lo bitrate mp3 encoders and see what comes
out... Maybe blend that back in with the original... It's not gaussian blur
but it's handy and REAL CLOSe in concept and application
|
Some time ago I wrote a moving window averaging algorithm for cleaning up
graphs of raw data. The algorithm could use various window functions including
Gaussian. Works as a low pass filter on audio files. I played with it a bit but
never
optimized it for audio use or characterized its [audio] response.
Later...
Ron Capik
-- |
|
| Back to top |
|
|
Chris Hornbeck
Guest
|
| Posted:
Thu Nov 03, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
On Wed, 02 Nov 2005 16:06:08 GMT, SSJVCmag <ten@nozirev.gamnocssj.com>
wrote:
| Quote: | Mr Hornbeck nailed this earlier there.. It's
|
It ain't me, Boss. Actually I don't believe that
a really good analogy can be made. Folks all talk
about "focus", "imaging" and such, and everybody
gets along famously, because we know it when we see
it, until asked to define the terms.
Already we have a definition for "blur" coming
from the math of Photoshop that differs in every
way from tried-n-true audio methods of generating
a "blurring".
In photography blur comes from mapping the 3D
world to a 2D surface, and the compromises needed.
This doesn't strike me as something with more
than the broadest conceptual links to audio
reproduction, idealized as point sources flowing
through time.
Great question, though, and very useful to ask.
Live another decade and we may even get a great
answer. Things happen fast...
Thanks, as always,
Chris Hornbeck |
|
| Back to top |
|
|
Les Cargill
Guest
|
| Posted:
Thu Nov 03, 2005 5:42 am Post subject:
Re: Gaussian blur for audio? |
|
|
Bob Cain wrote:
| Quote: |
Arny Krueger wrote:
In audio, this would be approximated by weighted samples taken near
the sample being blurred. The displacement would be in time, not space.
This is basically reverb or chorus.
No, a weighted average of nearby samples suppresses high frequencies in
ways determined by the weights, in general a lowpass. To get reverb and
chorus you add in weighted samples from a considerable distance in the
past.
Bob
|
What about slew rate limiting? Isn't that "blurring" in an
audio sense?
--
Les Cargill |
|
| Back to top |
|
|
Bob Cain
Guest
|
| Posted:
Fri Nov 04, 2005 12:46 am Post subject:
Re: Gaussian blur for audio? |
|
|
Joe Kesselman wrote:
| Quote: | Bob Cain wrote:
What about slew rate limiting? Isn't that "blurring" in an
audio sense?
Yikes! I guess you could call it that. I've never experimented with
it as an effect though, and wonder what it sounds like in substantial
amounts.
Should sound like a low-pass filter, right? Or am I misunderstanding how
you're using the term?
|
No, it is non-linear and doesn't have a lowpass type of model. A
lowpass will not introduce spectral components not in the input. Slew
rate limiting will. It clips the slope of the waveform rather than
clipping the magnitude. That may seem intuitively to just limit the
upper frequencies but it's a case where intuition fails.
Bob
--
"Things should be described as simply as possible, but no simpler."
A. Einstein |
|
| Back to top |
|
|
Arny Krueger
Guest
|
| Posted:
Fri Nov 04, 2005 1:16 am Post subject:
Re: Gaussian blur for audio? |
|
|
"Joe Kesselman" <keshlam-nospam@comcast.net> wrote in
message news:kvKdncW6usNPkffenZ2dnUVZ_tadnZ2d@comcast.com
| Quote: | Bob Cain wrote:
What about slew rate limiting? Isn't that "blurring" in
an audio sense?
Yikes! I guess you could call it that. I've never
experimented with it as an effect though, and wonder
what it sounds like in substantial amounts.
Should sound like a low-pass filter, right?
|
When slew limiting is severe, it reduces the high frequency
energy in audio signals.
But, while the consequences are something like a low-pass
filter, they are also different because as others have
pointed out, slew limiting introduces nonlinear distortion.
For example, a low pass filter won't create high frequency
IM, but slew limiting can.
how you're us |
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in