Top notch people in here.

@DrDave The 125 Hz point is the approximate peak of a rather broad boost I gave to that general region. Again, this makes sense only in the context of this particular recording. By boosting that area I also boosted the lower frequencies of the ambient noises. That increased the overall impression of the woman’s voice having more fullness.

In a sense this is pure trickery—I’m tricking the ear and brain of the listener into perceiving the spectrum of the sound in such a way that the voice of the woman being interviewed sounds—if not “normal”—than at least pleasant, and not causing listening fatigue.

Both you and @soundgh2 are right—the idea was to make it sound good (or at least better), and I’m glad that you and Ed (I hope) liked it.

I will try the two EQ notches--I did the roll off only. Tx. I would not have guessed 125 Hz since female voice only extends to 220 (unless you are a "tenor") but the results sound good and that is what matters!

@DrDave I’m for hire, if anyone’s interested. ;)

Here’s what I did. (The following pertains to the original audio of the entire interview that Ed sent me—not the snipped I uploaded earlier—although the approach was essentially the same).

First, I manually removed mic handling noises as well as some other intermittent sounds along with some low frequency bursts after plosives. I mostly worked with the spectrogram using the waveform only for reference.

Finally, I EQed the cleaned up audio: cut everything below 50 Hz, made a boost around 125 Hz to compensate for the lost lows, and made a dip at around 1.1 kHz. Then I made a few more surgical adjustments enhancing some frequencies for clarity and attenuating the most annoying ones.

I didn’t apply any compression, as that is something Ed should do according to his taste (and production requirements).

Don’t take this particular workflow as a template, though. Every cleanup job is unique and requires slightly (and, occasionally, drastically) different approaches.

THANKS ALL! @DrDave @Mark_the_Harp @soundgh2 I really appreciate you've given the time and effort. This generous help (and offers to help) has renewed my faith in online humanity :) I have indeed went with @Mr_Moore 's attempt and he's transferred me the whole track I needed to work with. So thanks again.

I recommend running a few passes so that in places where the noise overlaps--clunk+spike+frizz, you can fine tune the results. However, I recommend you go with @Mr_Moore unless something better turns up. I boosted the contrast slightly in the Vid as well just for fun. The big difference for me is taking out the handling noise. And here, unfortunately a lot of that noise is in the speech band. But you can see that most of it can be disappeared. I also tried deleting one channel and going with dual mono--which solves the phase problem as well--and I thought that sounded OK. The version above is the stereo version.

Changed phase and de-clunked, removed one big spike and some frizz. Password is audio I prefer the sound of @Mr_Moore--it is warmer! Some of the noise is now "lower", but it is still warmer. Excellent work! I did take the spike out, but I'm sure he could do that.

@soundgh2 It's entirely my fault. I shouldn't have used that cable in the first place. It was actually a shutter release extension cable that I thought would do the same job for audio. I mean, the levels on the GH2 screen were responding so I thought it was good to go (and I was in a rush). But now I've ordered that SESCOM cable that will allow me to monitor my audio from now on. Lesson learned... thanks again everybody.

@Mr_Moore Good stuff!! That is very acceptable! I'll have some background music so it'll mask it a bit.

@ed_lee83 The biggest problem is the reversed phase in one of the channels. Effectively, it cancels out the signal from the other channel when played back in stereo. Looks like mic cable wasn’t plugged in all the way. As a result, you lost a great deal of mids and lows.

I weaved a little bit of my magic over it. Take a listen and tell me if the result sounds satisfactory.

I have no objection to someone else doing this, but I think it would be cool to have four or five people upload different samples so ppl could see how the different techniques work. OTOH I'm wayyyyy behind on Grey's Anatomy......I think CEDAR would come out on top, so I would be happy to try Spectral Analysis

I've asked Ed enough questions about Lomos to do him a free rinse and polish :)