rms compressor acting strange

When rms compressor is applied to selected audio, only the very beginning of the selection is affected. The remainder of the phrase seems to be

unaffected. The lower the threshold, this phenomenon seems less noticeable.

version 3.0.4 / windows 10

The attack & release times could be set incorrectly for speech.

[ LevelSpeech2.ny does a good job of compressing speech ]

I don’t know anything about “rms compressor” but using compression on a selection could foul-up the attack and otherwise do “strange things” at the beginning & end of the selection where it suddenly kicks-in and kicks-out.

In this screenshot,
The upper track shows the original audio.
The lower track is after applying the compressor with the settings shown.
The entire second track has clearly been processed.


Please post a reply with an actual solution or don’t post at all please. Save the know it all snark for elsewhere. The point of a compressor is to compress what is above certain threshold, not the beginning of the audio only. If it does this then it is not doing what it is supposed to. My settings are very short to control vocal peaks.

OCENaudio has a native compressor which can be adjusted while the audio is playing.
[I don’t know if they have a forum which tolerates rudeness though].

My reply illustrates that on my machine, with the settings shown, the compressor is compressing the full selection.
Can you reproduce my test results?

Perhaps you could give more information about how to reproduce the problem that you are reporting.

I dont know what more to say but no matter what length of audio is highlighted, the beginning 5 to 20 ms (rough guess) is attenuated much greater than the remainder of the audio. It is vocal also but probably does not matter. I just wondered if at all possible there is someone who works at audacity or related that could tell me if this is a glitch that can be fixed or somehow intentional (cant see why). Uncheck compress based on peaks and make up gain boxes which from what ive read makes this compressor into rms mode. I just think the algorithm is not right.

The attack & release times could be set incorrectly for speech.

[ LevelSpeech2.ny does a good job of compressing speech ]

Thank you but level speech does not work well at all. I have it. It simply raises the volume level of everything without performing much compression at all. Exactly like compressor with “compress based on peaks” box checked. If I dont see different volumes become closer in level after using then it is not going to be acceptable to me.

LevelSpeech is capable of EXTREME compression.gif


FYI - Audacity also has a built-in limiter (which is a kind of fast-compression).

Or there are additional-optional [u]Nyquist[/u] compressors and limiters (in the Dynamics Processing section) and the “famous” [u]Chris’s Compressor[/u].

Or there are TONs of 3rd-party VST compressor plug-ins (although they don’t all work in Audacity). Compression of one of the “big-3” most-common effects so hopefully you can find one that works for you. If not, compression is probably not what you want. :wink:

Thank you. Ive used limiter and it works well. Compression is what I want. I just want a proper average (rms) of the vocal that im applying it to. The default compressor works well enough, I just have the extreme attenuation at the beginning of the phrase which seems like it should not happen. What i do now is simply use “amplify” effect and match it to the remainder of the phrase. I was looking to see if it was something I could avoid. One less thing to fuss about.

Are you applying the compressor to multiple tracks at the same time?