Peter Nielsen, you wrote:
>How effective is shareware noise reduction software in allowing the
>results from a cheap "noisier" mic to approximate those of a more
>expensive less noisy one? To what degree is this practical for
>regular use?
I've been involved with restoring recordings for many years. I have
used Digi's DINR and currently have the Waves Restoration Suite and
Bias Soundsoap Pro. I don't have the most expensive ones, Cedar and
Algorithmix.
I can tell you they all suck. Sometimes you can work magic, but it
takes a lot of effort in tweaking, and usually I hate myself in the
morning. It seems that concentrating on the noise makes me lose
perspective of what I'm doing to the sound.
The problems with noise reduction algorithms are 1) taking the life
out of the recording, and 2) artifacts. Both can be conquered by
mixing ambience back in after cleaning, creating an artificial mix
that would be suitable for production...but it would no longer be a
document of a natural soundscape.
For an example of typical artifacts, listen to:
http://ad2004.hku.nl/naturesound/dandugan/Barred%20Owl.mp3
Turn up the volume to hear the background between the calls.
-Dan Dugan
|