Parallel Compression on Vocals: How and Why to Do It
You may have heard of parallel compression before. It’s sometimes called New York compression.
Parallel compression is a way of making an audio track sound better by adding a little something to the main sound, but without messing too much with the main sound.
Does that make sense? I’ll try to explain more as we go.
People talk a lot about using parallel compression on drums and bass to provide a fatter sound.
However, I’m interested in applying parallel compression on vocals. This is because vocals are often the most important part of a track, and techniques for enhancing the sound of the vocal can be crucial.
So, in this article I want to look at why you might want to apply parallel compression to your vocals, and some ways that you can do this.
How Is Audio Compression Used in Music Production?
Before we look at parallel compression, we should start with regular compression and how it’s used in music.
We have looked at audio compression in music production in another article on the website, but I’ll provide an overview here.
An audio compressor is a bit like an automatic volume control. When the volume of the sound goes above a defined level (called the threshold) the compressor turns it down. How much it gets turned down is defined by another setting (the ratio).
With a low ratio, e.g. 2:1, the volume is turned down a little (to around half way between the threshold and the actual volume).
With a higher ratio, e.g. 5:1, the volume is turned down more (to around a fifth of the volume level that’s over the threshold.
So far, all this means is that the volume of the sound is turned down (and how much it’s turned down depends on the threshold and ratio settings).
Compression reduces the dynamic range of the audio. This means the louder parts of the audio have been squashed down closer to the quieter parts.
Now we can turn the volume of this audio back up to make up for the drop in volume (make-up gain). As you turn up the the volume you raise the level of the louder parts and the quieter parts together.
Now the peak level of the sound is back where it was, but the quieter parts are louder than before.
This causes the audio to sound more balanced, more present (closer to you), and maybe a little more intense.
In terms of balance, the level of the audio is evened out, making the audio sound more consistent throughout. This can make a big difference to instruments recorded live and, in particular, a vocal performance.
Parallel processing in audio production means that you have two copies of a piece of audio, and you apply different processing to each copy.
You can either make these copies by duplicating the audio, or by splitting the audio signal into two. In fact, you can have more than two copies, but we’ll stick to two here for simplicity.
Parallel compression is where the processing that’s applied differently to the two copies of the audio is compression.
Why is it Called New York Compression?
Just a quick note on why parallel compression is sometimes called New York compression.
Apparently, the reason this style of compression gained this name is because the technique was very popular with studio engineers in New York recording studios.
It seems that this compression technique was initially used with classical music recordings, where there the music can have a large dynamic range.
As a result it could be very difficult to hear the softer instruments in a classical piece. Parallel compression let the studio engineers make the softer parts more audible in the overall mix.
Parallel Compression on Vocals
So, what are the benefits of parallel compression on vocals?
In any vocal performance there will be parts that stand out due to their timbre, tone and volume. There will also be parts that aren’t so audible, again, due to their softer timbre and tone, and lower volume.
Parallel compression lets you raise the level of the softer parts of the vocal, making them more audible.
Using regular compression to do this could make the overall vocal sound flat due to the reduced dynamic range.
With parallel compression you have an uncompressed version of the vocal, which will sound dynamic and natural, mixed with a heavily compressed version that brings up the softer parts of the vocal.
Carefully blending these two versions can make the vocal sound much better due to the combination of these two vocal features.
Uncompressed Part of the Vocal Is Likely to Include Compression
There is something to note about the uncompressed version of the vocal here.
It’s very likely that what we are calling the uncompressed vocal or dry vocal is likely to have had some compression applied to it.
This compression could be applied as part of the vocal chain effects, for example (read more about vocal chain effects in this article).
So the uncompressed version of the vocal in parallel compression, just means that it is not compressed further during this process.
Parallel Compression in Ableton Live
We will have a look at some ways to do parallel compression in Ableton Live, although it can be done in all DAWs (as far as I’m aware).
We will start with the compressor settings that can be used for parallel compression, then look at some ways of mixing the compressed signal with the uncompressed (dry) signal.
As outlined already, the compressed vocal audio will be mixed with the uncompressed, or dry, vocal audio so you can use quite a lot of compression for this.
We are using the standard Ableton Compressor plugin here, which comes with all versions of Ableton Live.
Add the Compressor to the vocal track as an insert effect and dial in the settings below.
On the right of the Compressor you can choose the way the plugins responds to audio. The options are Peak, RMS and Expand. For this we will choose Peak, since this makes the Compressor respond quickly to short peaks in the audio.
Threshold needs to be set quite low so that compression will be applied to most of the vocal audio. A setting of -30dB will probably do the trick.
Ratio should be 4:1 or higher so that a fair amount of compression is applied.
Attack can be set to the lowest setting available so that the compression is applied to the vocal as soon as the level exceeds the threshold.
Release can be set to the lowest setting too, so that the compression stops acting on the audio as soon as the level drops below the threshold.
Dry/Wet should be set to 100% so that all you hear is the compressed signal.
Play the vocal and listen to the effect the Compressor has on it. Switch the Compressor on and off using the Device Activator in the top-left corner of the plugin as you listen.
These settings are producing about 18dB (-18dB) of Gain Reduction on the vocal recording I am using. Gain reduction is displayed by the GR meter in the Compressor plugin.
Apply Make-Up Gain
This amount of compression will make the vocal sound much quieter. Turn up the Output Gain until the level of the compressed and uncompressed vocals sounds about the same.
What you are doing here is turning up the overall level of the compressed vocal to the level of the original dry vocal.
But, you are also turning up the level of the quieter parts of the vocal so they are closer to level of the loudest parts. When mixed with the dry vocal, these quieter parts are now much more present and audible, providing the final combined vocal with more definition overall.
You could use the Makeup button to automatically add Make-Up Gain, but you will have more control if you adjust this setting yourself.
Mix the Compressed Vocal With the Uncompressed/Dry Vocal
If you listen to the vocal with this compression applied, it will probably sound quite unnatural, and not at all like what you are looking for from the vocal performance.
The next stage is to apply this amount of compression to part of the vocal and leave the other part uncompressed. Mixing the two versions is what provides the desired effect.
The dry, uncompressed vocal signal will sound natural, while the heavily compressed version adds extra definition to softer aspects of the vocal (by raising the volume of these quieter parts of the sound).
The impact of enhancing these softer, quieter parts of the vocal is one of the main things that makes parallel compression so effective.
Parallel Compression Using the Wet/Dry Control
The easiest way to apply parallel compression to the vocal is by using the Dry/Wet Control on the Compressor.
Not all compressor plugins have a Dry/Wet control, so you will have to check to see if the one you are using includes this feature.
The Dry/Wet control lets you adjust the balance between the compressed and uncompressed signals. At 100% only the compressed signal is heard, and at 0% only the dry, uncompressed, signal is heard.
So, as you raise the Dry/Wet level from 0% and 100% you will hear more and more of the compressed vocal mixed in with the dry vocal.
You will need to adjust the Dry/Wet control until the vocal sounds best to you. You could start by adjusting the settings in one of two directions, as outlined below.
You could start from 0% (only dry vocal) and add more compression until the vocal sounds good to you. What you are listening for is additional interesting things within the vocal (these interesting things being the quieter parts that you wouldn’t normally notice in the dry vocal).
Or, you could start from 100% (only compressed vocal) and remove compression until it sounds good. Here, you will be listening for the vocal starting to sound more normal and less artificial.
Whichever way you start (0% or 100%), when you reach the point where the vocal starts to sound good, you can make further adjustments to get the best sound.
This will be down to your own personal taste. You may be trying to reproduce the sound of a vocal you like on a commercial track, or you may just be trying to find a vocal sound that you like.
Alternatives to Parallel Compression Using Dry/Wet Control
Using the Dry/Wet control for parallel compression is the easiest way to do it. However, with this method, all that is available to you is the effect produced by the particular compressor plugin.
Some compressor plugins are modelled on analog hardware compressors, and these units can add saturation, distortion and other types of color to the vocal.
Others, like the Ableton Compressor, add fairly clean compression. This means that you will have to use additional plugins to add color to the compressed sound.
There are various options for including other audio effects with the compressor in parallel compression. Remember parallel just means mixing the compressed (wet) vocal with the uncompressed (dry) vocal.
For example, including a saturation effect for parallel saturation is a popular option. Saturation effects let you add harmonics to the vocal, which can make the vocal sound warmer.
You can read more about audio saturation and music production in another article on the website.
Since the saturation effect is applied to the vocal in parallel, mixed with an unsaturated version, you can apply high levels of saturation and mix this in for a more subtle effect.
The use of separate Dry and Wet tracks, Send/Return tracks and Effects Rack Chains allows you to add these additional effects plugins to the compressed version of the audio.
Although we will only talk about compression here, bear in mind that you could add other effects to the wet tracks, return track or effect rack chain alongside the Compressor.
Parallel Compression Using Separate Dry and Wet Tracks
To use separate dry and wet tracks for parallel compression, you need to copy the vocal recording onto a new audio track.
Add a new Compressor to the vocal on the new audio track and set it up as outlined above. If you already tried the Dry/Wet method above, you could just copy and paste the one you already set up.
For this method the Dry/Wet control should be set to 100%.
Set the volume level fader of the compressed wet track at zero, and the fader level of the uncompressed dry track at the normal level (i.e. the level you would normally have this set at).
Gradually raise the level of the compressed track until the vocal starts to sound good (or different). You are listening for an increase in the definition of the vocal performance. This will be due to sounds that are too quiet to hear in the dry track being more audible due to the increase in their level in the compressed track.
Alternatively, you could start with volume fader of the compressed track at full and reduce the level until the vocal starts to sound less compressed and more musical.
Once you have found the setting where the sound appeals to you, you can make fine adjustments to both the wet and dry levels to get the vocal sound you like best.
Something to note with this method is that you are adding an additional sound source (the wet compressed vocal) and this can increase the overall volume of the vocal.
Since a louder vocal tends to sound better anyway, you might think that the compression is improving the sound, whereas it’s just the increase in volume that’s doing it. You should make sure the overall volume of the vocal is adjusted to prevent this.
A good way to adjust the level of the two vocal tracks in Ableton Live is by using the Utility plugin. This lets you lower the gain on each of the vocal tracks without using the channel faders (which can help to avoid problems with any automation you might apply later in the mixing process).
Parallel Compression Using Send/Return Tracks
Send/Return tracks allow you to mix the amount of an audio effect that’s applied to a track. In this case, we can mix compressed and uncompressed versions of the vocal audio.
This means you don’t need a separate wet track with the vocal audio on it. All you need is your dry vocal track, and you send an amount of the signal to the return track (containing our compressor) to mix the signals.
In Ableton, you can make a new Return Track (just like adding a new MIDI or Audio track) and add the Compressor plugin to it.
You could add a new Compressor to the Return track and set it up as outlined above. If you already tried the Dry/Wet method above, you could just copy and paste the one you already set up.
For this method the Dry/Wet control should be set to 100%.
Just like the previous two methods, you will mix various amounts of the compressed wet vocal with the uncompressed dry vocal.
You could start off with the send control set to zero and gradually turn it up until the vocal starts to sound different and better, due to the increased definition of the quieter parts of the vocal.
As with the previous two methods, you could also start with the send level control at maximum and reduce it until the vocal starts to sound more normal and less compressed.
Again, you can make fine adjustments around this level to get the most appealing vocal sound from your point of view.
The use of a return track containing the Compressor (and maybe other audio effects too) can increase the overall level of the vocal.
If the audio level becomes too high it can introduce clipping and digital distortion, which sounds unpleasant.
If the level is only a little higher it can make the vocal sound better due to the increased volume (rather than because of the compression), and you don’t want this either.
You will need to reduce the levels to compensate for the return track audio. You could reduce the level of the uncompressed vocal, or reduce the level in the return track in order to reduce the overall level and avoid these problems.
Parallel Compression Using Ableton Audio Effects Rack Chains
Finally, we’ll look at parallel compression using Audio Effect Rack Chains.
As far as I know, this is something that is specific to Ableton Live, although you may be able to do something similar in other DAWs.
As the name suggests, in this technique you set up dry chain and a wet chain and mix the two to taste.
Starting with your dry uncompressed vocal track, drag in an empty Audio Effect Rack plugin (you’ll find this in Audio Effects -> Utilities in the Ableton Browser).
Before you can create your chains you need to Show the Chain List (even though you don’t have any chains yet) by clicking the little button that looks like it has three lines with bullet points.
A larger square saying Drop Audio Effects Here should appear. Right click in the square and select Create Chain.
Right click in the Drop Audio Effect Here square again, and select Create Chain to create a second chain.
You should rename the chains so you don’t lose track of which is which.
Right-click the top chain, select Rename and call it Dry.
Right-click the bottom chain, select Rename and call it Wet.
On the Dry chain, click the little speaker icon to mute it.
Drag a Compressor onto the Wet chain and set it up as outlined earlier in this article under Compressor Settings. If you have already tried some of the other methods here you could just copy and paste the Compressor that’s already set up.
For this method the Dry/Wet control should be set to 100%.
Turn the level of the Wet chain right down to -inf dB. Click the little speaker icon in the Dry chain to activate it.
Start playing the vocal track and gradually turn up the level of the wet chain. Again, you are listening for a change in the sound of the vocal as the quieter parts of the vocal become more audible, providing a greater sense of definition.
You could also start with level of the Wet Chain at maximum and lower it until the vocal starts to sound more natural.
Once the vocal sound is roughly where you would like it to be, you can make fine adjustments to achieve the best sound for your taste.
As you turn up the level of the Wet Chain you will find that the overall level of the vocal increases. You will probably need to adjust the level of the channel fader on the vocal track to compensate.
As outlined earlier, this is important to avoid clipping distortion and the increased volume making the vocal sound better rather than the compression.
I got these straightforward instructions for setting up Ableton Audio Effect Chains in an article on the MusicTech website if you would like to see more on the subject.
Give Parallel Compression a Try on Your Vocals
Making your vocal recordings sound good can be quite challenging.
Parallel compression can make a big difference to the sound of a vocal performance and is well worth trying.
Hopefully, the guidelines in this article are easy to follow and will help you to try this technique for yourself.
The examples used Ableton Live (since it’s a very popular DAW and is currently my favorite), but the same things can be done in other DAWs fairy easily.
One more thing to note is that you may want to add EQ into the process (you can read more about EQ in this article).
You could EQ the vocal before compressing it so that you are compressing something you are already happy with. Or you could EQ the compressed vocal, or the vocal that has been treated with parallel compression.
It’s all part of getting an impressive vocal performance from your recording, which will grab the listener’s attention.