Audio Expander acts as compressor not expander

  1. The audio expander - which is really useful for removing background noise does not function as it should, it acts like a compressor. It cuts louder signals instead of cutting softer signals and passing louder ones. Critical for my application!

Here is what the author of that LADSPA plugin says for the description:

As you can see, it is in some class of compression-type DSP and appears to be intended.

Here is related bug report for Audacity:

See the corresponding doc from the engine MLT:
https://www.mltframework.org/plugins/FilterLadspa-1883/#8

That parameter has a strange name because it is a status, read-only LADSPA port. Trying to set that does nothing, It is not clear what to do here.

Maybe consider removing the filter from Shotcut? I have never known what to use the filter for. And if it causes confusion for users then maybe it does more harm than good.

I could imagine a use for a “Gate” filter - which would accomplish what the OP is probably looking for. Something like this:
https://www.mltframework.org/plugins/FilterLadspa-1410/
or this:
https://www.mltframework.org/plugins/FilterLadspa-2602/

I am leaving it even though it does not satisfy this user’s needs.
Keep in mind that we only include SWH LADSPA plugins on Windows and macOS builds of Shotcut at this time.

Great to see such a full response!

An expander will allow sounds below a value that one sets to have a reduced amplification therefore they can be made quieter still. A gate removed the background noice and will often interfere with the beginning of spoken words but an expander with attack and release times can be configured to reduce background noise without causing this problem. But thanks for the link I will have a look at the noise gate quoted. Ian

I Googled this when the Expander wasn’t functioning as I expected either and thought I would add to @louisaian 's description if it helps. The existing Expander seems to cut off audio above its Threshold like a hard limiter and then the Attenuation is performed on the kept audio below its Threshold. I was expecting the audio above the Threshold to not be touched while the audio below the Threshold could have the Attenuation changed so that noise below a Threshold could be reduced. This could be because of differences with upward/downward expanders or intended use of this one. Thanks!