Re: [ecasound] Sidechaining a compressor with ecasound?

[Kai V. <kve...@ec...>]

Hi,

On Mon, 19 Sep 2011, S. Massy wrote:

>> yup, this works. Just make sure you feed matching number of channels
>> (e.g. a stereo stream to a LADSPA plugin with two inputs) and
>> ecasound will detect this and instantiate only one plugin and route
>
> Great! But how would this be achieved? Here is what I tried:
> ecasound -c -a:all -f:s24,2,48000 -a:1 -i raw/Vocals.wav \
> 	 -a:2 -i selfproc/lead3.wav \
> 	 -a:3 -i:loop,SCin -o:loop,SCout -eli:1427,0,75,100,-20,10 \
> 	 -a:1 -o:loop,SCin -a:2 -o:loop,SCin -chcopy:2 \
> 	 -a:4 -i:loop,SCout -epp:50 -o:jack,system

let's try:

  ecasound -c -z:mixmode,sum \
 	-a:sc -i scinput.wav -chmix:1 -chmute:2 \
 	-a:main -i imainnput.wav -chmove:2,3 -chmove:1,2 \
 	-a:sc,main -f:,3, -o loop,mixdown \
 	-a:mix -i loop,mixdown -eli:1427,0,75,100,-20,10 \
     	    -f:s24,2,48000 -o output.wav

So that should work. The scinput.wav can be mono/stereo thanks to 
'-chmix:1'. So we sort the three channels to right order, mix them down to 
'mix' chain via the loop object (it's important to specify the channel 
count for the loop object so it knows how many channels to pass through), 
run the LADSPA plugin, and write the first two channels of the result to 
output.wav.