naturerecordists
[Top] [All Lists]

Re: Filter Techniques -- Revealed At Last.

Subject: Re: Filter Techniques -- Revealed At Last.
From: Walter Knapp <>
Date: Wed, 23 Jun 2004 02:32:42 -0400
From: Marty Michener <>


>>>Nobody said it would be easy.

Too true.

When I first got into major filtering, it was for the frog CD. Before 
producing any of the stuff for that, I spent over a month just 
exercising all the available filters on the files. Trying various 
possible approaches. And tossing the results. At the end of that time I 
felt I could start. So, don't think you can just jump in, even with 
instructions, and everything will be quickly done.

Whatever software a person has, they should play with it. Don't get 
stuck in a rut in how you filter. Play with various ways of doing it. 
Never stop experimenting.

>>>Although I predict that, like several of my previous posts, not a single 
>>>person will read through all of this, let alone use any of it, I STILL 
>>>send all my hopes for good recordings and my best regards, ;^0

You are wrong there. I read all the way through most of your posts 
including this one, so there's one. I have gotten some ideas from your 
filtering posts. I am using mac and different software, so don't do it 
by the numbers as you tend to describe. But, my general methods are 
similar. I just translate it to my software. And add it to my own 
personal filtering methods.

I'm using software that allows me to do realtime preview of a whole set 
of filters. I can turn each one on and off and so on to check what each 
is doing. Adjusting settings as the filters interact. And set a sonogram 
at the end to monitor what it's doing to that. So, I don't make all 
those intermediate files. Once it's all right I can not only process the 
file in one pass, but save the filter set I used. I've found, however, 
that such sets cannot be applied to every file without modification. 
They more make a starting point for the next similar file.

I do use a noise filter. But very lightly. Those things mess things up 
easily. I use it at the end of the stack.

I'm not sure I agree with your idea of one heavy filtering. Like 
processing graphics, I find sound processing often can be improved with 
several lighter passes. And in any case, a heavy hand rarely does well.

I've also got a trick you don't mention. That's a dynamics filter. After 
you have managed to drag that ID clip out, then it can be used to remove 
or just subdue faint sounds caused by all that filtering. Set a level, 
it moves any sounds below that level by a specified number of dB. 
(that's a highly simplified description) Helps in cleaning up heavily 
processed files. Though it has to be applied very carefully, calls have 
a considerable dynamic range, and removing the quiet part, even in the 
middle of the loud part, can result in surprising changes in the call.

Worst was when I was doing the CD and would use the dynamics filter I'd 
supply listening tracks with and without the filter. Most could not pick 
out what the change was. But i could hear it. In some ways that's the 
way good filtering should be, unnoticeable.

And the very best way is a quality recording to start with. Filtering 
cannot substitute for good technique or good equipment. If it could, 
this would be a much cheaper game. And of course there is the luck 
factor. That magic time when everything onsite is perfect.

Walt






________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU