In message , Jim Lesurf
writes
In article op.xmae0omdg15l5m@jamesm6700, James Perrett
wrote:
On Mon, 08 Sep 2014 03:16:03 +0100, Johny B Good
wrote:

On Sun, 07 Sep 2014 16:44:21 +0100, Jim Lesurf
wrote:
[big snip]

I've skimmed through this thread and my first impression is that most of
the developments in the past 15 years seems to have gone by unnoticed.
One of the fastest ways to find clicks is to look in the frequency
domain (which is what Jim seems to be doing)

Not quite. I've been looking at using differentiation. That does of course
'slope' the frequency response to emphasise HF and reduce the impact of LF.
So is aiming at the same kind of distinction as...

but the newer versions of Audition include a frequency/time view as
standard so the worst of the clicks are obvious.

That's a nice idea. I'd thought of it but it would involve far more in the
way of programming from my DIY POV. FWIW I've also been wondering about a
way to 'plug in' to Audacity some method like this. But I don't know enough
about Audacity as I've only really been using it recently. I've also
wondered about having Audacity show added tracks which are the
differentials to make the process easier.

I have the impression that the software you suggested, though, isn't for
Linux and isn't open source. So isn't particularly attractive for me. As
much as anything I'm wondering about this from the POV of examining results
versus method in detail.

I'm no expert, but I would have thought that you need a program which
would look at (say) 7kHz for clicks, and note the times when they
occurred. It would then replay and re-record the audio, using the timing
information to blank out the clicks from just before they occurred to
just after. A refinement would be to take a sample of the audio at just
before and just after a click, average it, and use it to fill in the
blanked-out bit. Is this the sort of thing that is done?

--
Ian