naturerecordists
[Top] [All Lists]

Re: Re: Elk bugling

Subject: Re: Re: Elk bugling
From: Aaron Ximm <>
Date: Mon, 30 Jan 2006 13:53:48 -0800
Brit also shared a slightly longer sample with me, which I took a stab
at applying Samplitude's native noise reduction filtering to. This
isn't a straight frequency filter; you provide a noise sample
(typically a selection of the source file) and it attempts to subtract
that sample in a tiled fashion from the original.

It can be very effective for stable noise (hiss, hum, etc.) but you
have to play with the various paramters to avoid the classic  "rounded
FFT step" noise reduction artifacts. I can hear digital artifacts in
these results, but I nonetheless find it an improvement...

...but it's a bid hard to tell since they are present in the source
MP3 as well -- and the calls themselves have such an odd sound
themselves. :)
 http://www.quietamerican.org/download/dropbox/Cervus_elaphus_cleanup.MP3

For reference, here's the original:

http://www.quietamerican.org/download/dropbox/Cervus_elaphus_Demo_Long.mp3

I played with applying additional parametric EQ to the post-NR file
but found myself taking the remaing edges off without making a
dramatic improvement to the remaining noise floor... but YMMV. :)

 best,
 aaron


> > I put a small clip of my elk bugling recording up on the site.  I
> > think the hiss in the background needs to go, but I am having troubles
> > editing it out.  I tried using the parametric equalizer, as suggested,

--
  
  www.quietamerican.org

  83% happy
   9% disgusted
   6% fearful
   2% angry


________________________________________________________________________
________________________________________________________________________

<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