Detecting noise reduction on a spectrograph

[hvbias]

Noise reduction like CEDAR is one of the banes of digital audio. It "sucks out" the sound of the silence between notes creating this dead, sterile sound. Very hard to describe it better than that, maybe someone else can try. I have always relied on listening to music to hear it, but I'm wondering if there is a more objective way to do this, for instance on spectrograph analysis?

 

[Blumlein 88]

Yes, a spectrograph should show that. The one in Audacity for instance you can set the lower threshold for the level at which a signal shows up. If the CEDAR process is rather obvious it should show such a thing. Especially if it drops to near silence at low levels like a gating process. I think the spectral editing part of CEDAR can cause what you are referring to and that should show up.

I've no experience with CEDAR processed tracks. Do you have one you could share 30 seconds of for us to see what it looks like in the spectrograph?

Now if someone does a good job with it you likely wouldn't hear anything remiss and seeing it in a spectrograph might be hard.