>> > What about Spike Cleaner which is disabled? We want a better
>> > Click Removal, but is Spike Cleaner a potential answer?
>> No, it's terrible code.  A bad place to start from.
>
>Why not rip it out then instead of having translators translate it?
 
+1

 
Peter Sampson
Tel: +44 (0)1625 524 780
Mob: +44 (0)7732 278 299
From: Gale Andrews <gale@audacityteam.org>
To: audacity-quality <audacity-quality@lists.sourceforge.net>
Sent: Wednesday, April 18, 2012 6:54 PM
Subject: Re: [Audacity-quality] Removal of CleanSpeech from our code


| From James Crook <crookj@indigo.ie>
| Wed, 18 Apr 2012 10:28:28 +0100
| Subject: [Audacity-quality] Removal of CleanSpeech from our code
> On 17/04/2012 23:16, Gale Andrews wrote:
> >> The way forward, I think, is to create a wiki page with a title like
> >> 'Proposal Cleanspeech Cleanup'.
> > In that case should it be called Cleanspeech Cleanup and
> > Chains Rationalisation?
> No.  It doesn't need the longer name, any more than it might have needed
> to say 'and Truncate silence fixes'.
>
> At the level of cleaning up mess from cleanspeech, there is good chance
> of getting developer interest.  I doubt you'd get a developer interested
> in open ended improvement to chains.  The way to get what I think you
> want is to question 'band aid' fixes.  A proper fix to the parameter
> mechanism is likely to be cleaner, and better and less work than lots of
> sticking plasters.

I'm +1 on global improvements to the parameters mechanism,
without making chains improvements "open ended". Just not
sure the proposed title conveys that if you were looking at
the list in the Proposal Category. 

You or I could easily step up and complain that the chain
commands are not translated, too. 


> > What about Spike Cleaner which is disabled? We want a better
> > Click Removal, but is Spike Cleaner a potential answer?
> No, it's terrible code.  A bad place to start from.

Why not rip it out then instead of having translators translate
it?


Gale


> >> Saving and restoring of noise profiles needs some thought
> > If so, then so do the other effects which supported CleanSpeech
> > presets? That then begets a comprehensive answer for all
> > effects?
>
> The key problem with noise profile is that unlike most effects (which
> have a few text parameters) it has a large number of non-text parameters
> - and so it has a special case mechanism.  Special case mechanisms cost
> us time and energy.  We should be looking for one mechanism for storing
> parameters, whether an effect or other step, and whether the step has
> many or just a few parameters.
>
> --James.



------------------------------------------------------------------------------
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring today. Free.
http://p.sf.net/sfu/Boundary-dev2dev
_______________________________________________
Audacity-quality mailing list
Audacity-quality@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/audacity-quality