sinusoidal synthesis (was Re: [Clam-devel] using eclipse

Pau Arumi parumi at iua.upf.edu
Wed Jul 4 08:11:36 PDT 2007


En/na Greg Kellum ha escrit:
> Hi Pau,
> 
> Feeding all the ports to the SMSSynthesis does seem to be mandatory.  I 
> tried it myself using the streaming SDIF example, but 
> CanConsumeAndProduce always returned false.  (I think.) 

You're right (and I was wrong) you can leave outports unconnected
but not inports.

Then my advice is to generate a dummy feed. And leave for Later the 
  good solution.

pau




> produce any audio.  Then, I tried it with the last release version of 
> the NetworkEditor, and it sounded awful although it did produce audio.  
> Then, I tried it with the current developer version of the 
> NetworkEditor, and it only produced an annoying beeping noise and 
> printed the message "Warning: 1 sinks were not fed, so could not send 
> audio to the callback." to the command line about one million times.
> 
> So, it looks like if I want to do just a sinusoidal resynthesis, I'll 
> have to make a new class that is similiar to SMSSynthesis.
> 
> Best,
> Greg
> 
> On 7/3/07, * Pau Arumi* <parumi at iua.upf.edu <mailto:parumi at iua.upf.edu>> 
> wrote:
> 
>     En/na Greg Kellum ha escrit:
> 
>      > By the way I also wanted to ask you how the best way to go about
>     doing a
>      > sinusoidal resynthesis (with no residual) would be in CLAM?  I would
>      > guess that one would have to create a new composite -- something
>     like
>      > SinusoidalSynthesis -- that would call the SpectralSynthesis class,
>      > handle the overlap and add, and handle the phase alignment, no?
>      >
> 
>     Just use the existing SMSSynthesis processing. Feeding all its
>     in-ports is not mandatory.
> 
>     pau
> 
>     _______________________________________________
>     Clam-devel mailing list
>     Clam-devel at llistes.projectes.lafarga.org
>     <mailto:Clam-devel at llistes.projectes.lafarga.org>
>     https://llistes.projectes.lafarga.org/cgi-bin/mailman/listinfo/clam-devel
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Clam-devel mailing list
> Clam-devel at llistes.projectes.lafarga.org
> https://llistes.projectes.lafarga.org/cgi-bin/mailman/listinfo/clam-devel





More information about the clam-devel mailing list