As you begin to record in your home voiceover studio, you’ll start hearing about “Compression.” Compression is one of the least well understood audio tools available to us.
During one of the first VO classes I took, another student was explaining their rule for recording. They said, “…you have to use compression” Emphatic. No room for discussion.
This struck me as kind of strange. I’d used Compressors in the recording studio – real ones with dials and knobs and stuff – and had a pretty good idea how they worked and where they normally got used. However, I was struggling to understand their use in VO. If voiceover was supposed to be “conversational”, “real” and “non-announcery”, then how did Compression help that? Compression tends to make things punchy and processed sounding – like a radio DJ voice.
Using Compression? Be careful!
The first thing to realize is that Compression, like any other complex audio effect, cannot be “undone” once applied to the audio. As with any effect applied to your audio, if you aren’t sure how to use it, it’s probably best not to.
A simple gain adjustment – something altered using “Amplify” for example – can likely be fixed or adjusted by a client without impacting the audio quality. But, if we use a complex process like Compression on our audio it can definitely cause problems further on in the production process.
If we hand off a file which has been Compressed (or Limited, or even had Noise Reduction applied), that has become part of the audio in a way that cannot be removed. (Even EQ can cause this). When a client asks for “Raw” audio, they are most likely requesting that you do not use any type of “dynamic Effect.” Compression is a dynamic Effect by definition.
How does an audio Compression Effect work?
Compression is triggered by the recorded audio reaching a certain volume level. This is the “Threshold” and can be adjusted directly on the Effect. Usually the settings on Compressors are for the incoming Loudness from the recorded track. They will attenuate audio as the loudness goes above that Threshold. This means that if the volume is below the threshold set on the Compressor, it will not have any effect upon the audio. You can think of the Threshold as a thermostat on your home HVAC system. Nothing occurs until the device kicks on.
Once the Compression Threshold is reached, the Effect begins reducing any additional increase of volume by a specific amount. This is the Ratio of Compression.
A low ratio of 2:1 lets one decibel through for every two that enter the Compressor. A higher ratio of 10:1 lets one decibel through for every 10 that arrive. With a high Compression Ratio, once the volume of your audio reaches the threshold, it hardly gets any louder at all which can sound quite artificial. A Compressor with a very high ratio starts to behave a lot like a Limiter.
Compressors have “Knees”?
This weird phrase which is used on many Compressor controls refers to the transition area of the Threshold. The “softer” the knee, the more gently it transitions through the Threshold point. A “sharp” knee is makes the action of the Compressor very abrupt and noticeable. On the other hand, if the angle of the knee is too soft it might actually start working before we really want it to do so.
Compressors do not make things louder – so, we need “Make Up Gain”
This is where a Compressor differs from a true Limiter. With a Limiter, we are increasing the volume “into” the Effect. That lets us Limit the Peaks with a high degree of accuracy.
By contrast, Compression will only make things quieter. Using the combination of Threshold and Ratio, a Compression effect only reduces volume in our audio. That is how it gets its name – it compresses the dynamic range and loudness. If we then want our resulting audio to be louder, we need to use a “Make Up Gain” adjustment after the audio is compressed. This amplifies the compressed audio we just created.
Because we’ve limited the dynamic range of the recording, and then made everything louder, the resulting audio will tend to sound “punchier”. This can easily be overdone and is often not desirable if anyone else will be doing any type of processing to our audio. It’s generally a bad practice to use multiple passes of dynamics tools on audio.
Attack and Release – Don’t be lagging!
Attack and Release values control the speed with which a Compressor reacts to the incoming volume (Attack) and how quickly it stops working (Release). For voiceover recording, we generally want a quick attack and release. If a Compressor reacts too slowly to the audio we’ll tend to notice the effect taking place – it can sound like our vocal track is suddenly getting squished. Not all Compressors allow you to control these variables.
Coloration, Saturation, Distortion – oh my!
As with preamps (see the section “Can Distortion Sound Good?” in my resource on audio interfaces), certain types of Compressors were historically used because the distortion or sonic coloration they added to the signal was pleasing within a musical context. Often described as a “Classic” effect, these are generally less helpful in a voiceover recording context, where transparent sound and clarity are key.
Putting a Compressor to work in your home voiceover / podcast studio
When we use a Compression effect on our audition, it’s important to ask whether that effect might be overwhelming the performance. Do we hear the processing rather than the performance? That’s a good indicator that we may be doing to much. With Compression, it’s easy to fall into that trap.
In my voiceover recording classes, I talk about Compression as a “tool of high leverage”. That’s a nerdy way of saying “…when you start to lean on a long crowbar – it doesn’t take too much pressure to start breaking things.”
Last week I went over the controls found on most Compressor plugins that we’re likely to encounter. Before we start applying them, it’s worth asking our client if it’s even necessary at all. If another person will be processing your audio further in the workflow, using compression can seriously impact what they might be able to do later on. That has a way of frustrating clients. If we are delivering final work to a client project, best practice is to confirm that applying compression is OK with them.
Some clients will want to handle that themselves, while others may expect you do it. It’s also possible that your client may have no idea what you are talking about. But it’s important to have that conversation.
What do we want a compressor to do in the VO studio?
I talked about two key ideas:
- Compression is activated when your audio reaches a “Threshold” level – a triggering amount of volume. Until we hit the Threshold, nothing actually happens.
- Multiple passes through dynamics tools is generally a bad idea. Use Compression once.
Used judiciously, Compression can solve several problems for us:
- It can smooth out performances which might have a bit too much volume variance (a wide dynamic range).
- The “make up gain” can boost the loudness of our audio so it’s appropriate for the project.
- Constrain dynamics with a bit more subtlety than what we do with a Limiter.
- It can sometimes “heal” raw takes that Peak past -0 dB
One of the reasons we try to control the dynamics is to provide a better experience for the listener. If our volume is reasonably consistent, then they don’t have to keep adjusting the playback level when listening to your audiobook or podcast. The trick here is to tune that “Threshold” so that the quieter portions are not affected. If we set the Threshold too low, it will cause everything to be at a similar volume, which gets noticeable quickly and sounds quite artificial. At that point, the effect starts to speak more loudly than the words we are saying. I realize that’s a bit of a grey area. But if we’re noticing “how” something sounds rather than the content, that’s a good test that we’ve been a bit too aggressive.
Dialing in your voiceover compressor settings
I’ll usually play with the Threshold first – that lets me dial in when the Compression starts to impact the raw audio. Depending upon the user interface of your specific effect, this can be a needle which starts to move, or a change in color of the level meter. I find that Apple’s AUDynamics Processor provides a very good interface to see this occurring. It changes from blue to red as you pass the Threshold point.
Once I’ve adjusted the Threshold, I’ll decide how aggressively I want the Compressor to compress. That’s the Ratio I talked about last week. A ratio in the 2:1 range is usually pretty transparent, which is what we’re aiming for. Going past 4:1 is often very audible. If you are using the AUDynamics Processor (shown above)it just shows a curve rather than giving a ratio, but again it’s a very intuitive interface in that a flatter curve means that less audio is getting through. That’s the equivalent of a high ratio.
After that, I’ll either use the Make Up Gain to bring everything up to the desired level, or use a separate gain adjustment (Amplify or Normalize) to tune in the delivery volume I need.
Compressors I like for a reliable voiceover workflow
Apple AUDynamics Processor
On the MacOS side of things, there’s Apple’s native AUDynamics Processor. This comes with every Apple computer and might be the only effects which appear in your recording software. It’s actually a very useful tool with a few other tricks. But, it provides a very neutral, clean compression with a very intuitive user interface. As you play your audio through the Dynamics Processor, you’ll see how adjusting the Threshold point activates the compression. Then you can adjust the reduction curve for more or less of an effect.
HoRNet Plugins Dynamics Control
The HoRNet Dynamics Control acts in a similar manner to Apple’s AUDynamics Processor, so those of you on Windows systems won’t miss out (there is a MacOS version as well). You have a graph-like interface to view the audio levels and can set the Threshold and curve in a similar manner. I’ve found this bundled free through some promotions in the past, but these days it seems to be available for a nominal fee.
Kotelnikov from TDR
Another amazingly useful (incredibly free!) plug-in from the folks at Tokyo Dawn Records (their NOVA EQ is on my “must-have” list). The Kotelnikov Compressor‘s user interface here is a bit more opaque – mimicking a hardware rack style compressor. But all the controls are easy to understand, and it shows how much compression is being added through a quick-reacting display.
Waves Compressors
Waves makes a bunch of popular compressors, but most are designed to add “character” to the signal – which is not preferred for voiceover. If you stay away from the more “classic” settings, these can work well – Wave’s CL2A – Renaissance Compressor / R Vox Compressor
Melda Productions MCompressor
(big tip o’ the mouse to George Whittam who introduced this to me)
I have to say that I was not familiar with Melda Productions plug-ins until just recently. George spoke about this in a session I attended and it’s a solid choice with some good features. Nice and neutral sonically and a price that is hard to beat (that would be free).
Good “Native” Compressors
Presonus Fat Channel
If you are a Studio One user, this factory-supplied Channel Strip contains a good simple compressor. It also appears in the Universal Control for the Revelator io24 and can be used as inbound processing for your recordings. I find myself wishing they would make this available outside of the Presonus ecosystem, as it has easy setup and a clear interface.
Adobe Audition
Adobe Audition supplies a versatile Dynamics processor which shows the audio in a graph similar to the MacOS Dynamics Processor. They also provide a “Single Band Compressor” which can be useful, though the user interface does not supply much information.
Audacity Compressor
Audacity supplies a very workable compressor with their proprietary Effects. Though they do provide a rough graphic which corresponds to the controls, it’s a bit cryptic. This native compressor also turns on the Make-up Gain by default, and boosts the Peak output to -0 dB. You can uncheck this option, which I find to be a better approach for voiceover workflows. In most cases, you want to control your delivery loudness.
The Big Question – Should we compress?
The cold truth is that I usually try to talk folks out of using compression for most voiceover applications. I’ve always been a bit less likely to bring it up as an option – especially with voice actors who are learning studio skills. The main rationale is that while it can “sound like” audio we might have heard, it’s far too easy to overuse the effect.
Used inappropriately, we lose the nuance of performance dynamics as we push our audio into a narrow volume range. The tool we use becomes very apparent and can easily overwhelm the performance itself.
Many audition specs often give direction such as “non-announcery”, “conversational”, or even the dreaded “don’t sound like a voice actor…” When we layer on compression, it tends to make things sound the opposite of those guidelines.
Certainly, if a client asks for “RAW” audio as a deliverable, that’s usually followed by the example “No Compression!” Compression reduces the dynamic range of our performance. Once compression is applied to the audio, it can severely impact what the “next” person can do to our audio.
A simple High Pass Filter can reduce rumble, it shouldn’t interfere with further processing. I have no problem applying that type of EQ to a file and still calling it “RAW”. But dynamics tools – Compression or Limiting – should really only be used if the client signs off on it. As I mentioned in last week’s article, it’s easy to make things sound overproduced with careless compression use.
Paying attention to the Threshold is a key place to begin. That makes us ask exactly why we are reaching for compression in a given situation. This more surgical application of this powerful effect is key to setting things right.
As I was writing the resource above, it struck me that a companion video might be helpful, so you could hear how some of these settings change an audio recording.
Compression for Voiceover – 3 Key Settings
More videos available on YouTube/@JustAskJimVO
This home recording studio resource expanded on a few of the weekly posts I share through my “Tuesday VO Tech Tips” – Each week I send out a new article about using recording technology in your home voiceover studio, and how to balance those technical challenges with the need to be creative behind the microphone. You can find archived copies those posts in the resources section of this site.
If you would like to receive those resources via email on the day they publish, please take a moment to share your contact information through this sign up form.
Thank you!
