I’m looking for an effect to compress spikes (most often when a lavalier mic gets snorted into ). At the moment, I’ve been using the inbuilt compressor, but I’ve noticed that even when the decay is set to 1 second, it still compresses the following audio unacceptably. These spikes generall hit 0db, whereas the rest of the audio sits around -10db.
What I guess i’m looking for, is an effect that will compress/limit such that the spikes get clipped/muffled acceptably (not just a bare digital clip).
Can anyone point me in the right direction–I’m not sure the exact effect I need.
The best solution is to avoid snorting into the mic.
If the microphone gets blown on, this will usually cause a big surge in low bass frequencies. Using a high pass filter set to around 100Hz will help to reduce these sounds without having much affect on a speaking voice. If there are just a few loud “plosives”, you can zoom in close and select the offending peak and use more severe filter settings (steeper slope and slightly high frequency).
Peaks in audio can also be limited using a “limiter” effect. I think a “hard limiter” is included in the standard Audacity 1.3.12 effect menu.
Limiters are designed to only affect the audio where the peak amplitude exceeds the threshold level. It is a very basic, some would say brutal, tool that essentially hacks off the top/bottom of the waveform. The “hard limiter” effect can be softened (made less brutal, but also less effect) by increasing the “residual” setting.
Sounds exactly what I’m looking for (given that it’s a snort & nothing else is audible, all that I need is that it’s less loud). The reason I ask is because I usually try to take these peaks out as my first processing step & then apply a 2:1 compression on the remaining audio. In the interests of quality, I just wanted to make sure it didn’t fiddle with the audio below the threshold…
Thanks so much.
“Wet/Dry” is fairly standard terminology with effects.
“Dry” means the original, unprocessed sound.
“Wet” means the effect/processed sound.
Many effects allow you to adjust the proportion of original (dry) sound to the amount of processed (wet) sound.
“Residual level” is a made up term for this effect.
“Hard limiting” a sound chops the tops/bottoms of the waveform flat - this produces a nasty sounding digital distortion. The “residual level” control appears to allow a proportion (a residual amount) of the peaks to remain, so the peaks are not cut off flat, but just “shrunk”. The more “residual level”, the less the peaks are shrunk, so the less harsh (weaker) the effect is.
). At the moment, I’ve been using the inbuilt compressor, but I’ve noticed that even when the decay is set to 1 second, it still compresses the following audio unacceptably. These spikes generall hit 0db, whereas the rest of the audio sits around -10db.
)