Attack and Release Times for Compression

Question from Chip:

I enjoyed your episode with Emily Prokop and learned some new tricks to use with Reaper.

I just got Scheps Omni Channel and I love it.

I’m trying to find out more about how to set attack and release times for compression. I’ve heard people use language like “fast attack” or “slow release” but I don’t understand how that translates to milliseconds.

Do you have a reference that lists a range of times for the different attacks and releases on a compressor?

I’ve asked the same question a few other places and I haven’t gotten a good answer. So I’m starting to suspect I’m coming at the issue from the wrong angle.

My current setting is an 11ms attack with a 110 ms release. I got it from an NPR guide and it seems to work.

Also, I recently learned that SWS extension will let me normalize audio to specific LUFS levels on reaper. To level my audio I normalize the whole thing to -24 LUFS then I put splits in after each phrase and normalize each clip to -24 LUFS. This results in a nicely leveled file.

I know this probably causes my noise floor to bounce around but I’m not sure it matters because I use an expander. If I add more than like 5 dB to a clip I add RX Voice De-noise just to that clip and adjust it so the noise sounds similar to the clip before and after it.

After everything is leveled to -24 LUFS I use a limiter to boost it up to -19 LUFS.

This seems to work for me but I don’t want to do something that would be considered bad engineering. I just do dialogue using the clip by clip method. I don’t use it on music.

I hope you are having a great weekend.

Answer from Chris:

Hi Chip! Good to hear from you.

Sounds like you’re doing just fine 😉

The attack time of a compressor will determine how much of the initial transient gets through before the compressor kicks in. Perceptually, this will allow the transients to “poke out” a bit more (longer attack) or not to poke out as much (shorter attack). Roughly, I would consider 0-10 ms very short, 10-35 ms medium, and 35+ ms long.

The release time of a compressor will determine how long it takes the compressor to completely deactivate after the sound goes below the threshold. Perceptually, and subtly, this will make the voice sound like it’s closer (shorter release) or further away (longer release). Roughly, I would consider 0-100 ms short, 100-500 ms medium, and 500+ ms long. Also, most compressors are capable of up to a 5000 ms release time; generally 2000-5000 ms release times are used to broadly level out a signal which is loud for a while, then quiet for a while, etc.

Also, normalizing each clip after you split up the original track is probably only necessary if the recording is extremely dynamic, meaning very loud at times and very soft at times; this is rare, but can happen when someone has terrible mic technique. If the original track is OK in this regard, you’d probably be OK just using a compressor or two on the entire track, which would save you time of splitting them up and normalizing separately, etc.

Hope this helps.

Want to receive the Daily Goody in your email, daily or weekly? Subscribe free here.

And please keep in mind, the Daily Goody is only a tiny little tip, fact or lesson everyday. Please don’t expect any of these posts to be long, earth-shattering masterpieces that instantly answer every single question you can think of and completely transform you into a world class podcast engineer. “Little by little, a little becomes a lot.”

.