Final Cut Pro X is an excellent video editor, but how does it stack up when mastering audio? When we utilize interviews in our videos, we want to ensure our speakers sound crisp and brilliant. With FCPX’s powerful Logic audio plugins, this is possible with a few simple tweaks.
Wait, you can master audio inside of Final Cut Pro X?
At 9miles Media, our editor of choice is Final Cut Pro X. Many of the videos that we produce feature interview soundbites overlaid onto music and B-roll footage. While Final Cut Pro X is an incredibly powerful video editor, it hasn’t always been a great audio mastering tool.
We certainly could roundtrip audio to a dedicated mastering program, but why go through all the hassle? Our Content Production Specialists have found that the Logic audio plugins in FCPx are actually very powerful (that is, when you know how to use them). After years of experimentation, we’ve discovered a simple process that allows us to use FCPx to its fullest sound potential. Read on to find out our process.
The Goal
What does the term “mastering” actually mean?
Let’s say we’ve got ourselves a short video project. In this project, we have interview footage— with B-roll overlayed—and a music track sitting underneath. Typically, you’d just increase the levels of the vocals and decrease the music levels until you can hear the vocals pretty well over the song. This can work. But is it ideal?
You’ll notice with this process, that vocals never really “sit” well in the entire mix. They can often sound muddy and muffled—like the speaker is under a blanket. You’ll also notice you can only increase the levels so much before the vocal track starts to clip (when your audio meter hits 0 dB and starts to distort).
Take a look at the following example from a recent video project. Jay, one of our Content Production Specialists, is talking on camera with a song playing underneath. The vocals have been increased as much as possible before reaching 0 dB on the audio meter. No audio effects have been implemented. Take a listen!
How does it sound?
The vocals don’t quite sit well in the whole mix. They sound like they’re sitting behind the song and are difficult to hear.
But the vocals have been increased as much as possible! How can we fix it?
This is where mastering comes into play. To answer the initial question, mastering is when we balance all elements of a mix to create a uniform and cohesive feel. Like everything belongs together, and all elements of a mix get along well. To do this, we’re going to use three tools in FCPX: EQ, compression, and limiting.
EQ
Read this explanation here.
Then, download this handy cheat sheet.
If you didn’t read the explanation and want a TL;DR: EQ is used to boost or reduce the levels of different frequencies in an audio signal. Basically, we can reduce the bad and increase the good. (This isn’t a substitute for a poor recording, however!)
Step One: Sweep for Bad Frequencies
Along the entire frequency range, there exist certain frequencies that just don’t sound right. It might be reverb in a room, or weird resonance and ringing in the voice. Our goal is to reduce these bad sounding frequencies. So, how do we find them?
Add a “Channel EQ” to your audio channel (I like to double click into my multicam edit and add it directly on the master track—I would highly recommend this so you don’t have to copy and paste effects.) This will be your first EQ.
Grab one of the parametric bell filters (the ones in the middle), and make it as narrow as possible and increase the level all the way up. Now, play your track, and slowly sweep up the frequency spectrum as seen here:
As you sweep up and down the frequency range, you’ll notice very peculiar sounds coming from certain frequencies. They may sound very reverb-y or have a bad sounding ring to them. It is especially apparent in the 100–300 Hz, 800 Hz–1 kHz, and 2–3 kHz range.
Once you’ve identified these frequencies, cut the levels around -12 dB on them as seen below:
There were four frequencies I identified and cut.
Step Two: High Pass
Below 100-150 Hz (depending on the voice), there can exist a lot of signal that is muddy and unneeded. Let’s get rid of that.
A high-pass is what it sounds like—it allows the high frequencies to pass through. Add a second “Channel EQ” to your audio track. Depending on the voice will depend on what crossover you set your high pass at.
As seen in the image above, Jay’s voice sounded best when I set it right at 85 Hz. Too high, and he starts to sound thin and tinny. Too low, and he starts to sound muddy again.
Step Three: High Shelf
Been looking at that cheat sheet? Good! If not, pull that back up. Right around 3 kHz and up, vocals can greatly benefit from a boost in this area to increase clarity and brightness. To boost this, we use a shelf. That’s the purple icon in your Channel EQ.
Play around with the starting frequency of your shelf. Mine is set at 5300 Hz for Jay. The farther down you set it in the frequency, the more apparent the effect will be. Most vocals can benefit from this top end boost.
Step Four: Cut and Boost
Our final step in the EQ process is to cut and boost certain frequencies along the middle spectrum here. Use the cheat sheet as your guide. Typically, I like to boost around 180-220 Hz and 1800-2200 Hz to increase clarity. I like to cut around 300-500 Hz to reduce muddiness and boominess.
Here is the end result:
Compression:
I recommend reading this handy reference here.
Compression helps normalize the audio levels of a recording. When used subtly, it helps even out the voice and makes it clearer to understand. However, if you overuse it, it can make the voice sound unnatural (think radio voice—we don’t want that). The human voice is so pleasing to listen to because of the natural ups and downs of volume and intonation. We want to keep that.
Apply a “Compressor” filter onto your track. Zero out your threshold, ratio, attack, and release. I’ve set my vocal track to 0 dB in the timeline as well. I’d recommend doing this as we’ll set the volume with the limiter at the end.
- Set your attack to 1-5 ms, and your release anywhere from 15-30 ms.
- Set your ratio around 1.7:1 to 2.2:1. This is subtle but still effective.
- Play your track and slowly increase your threshold until you start to see the compressor working. I like to set mine until I see around a 5 dB reduction.
See my settings below:
This reduces peaks and increases the quiet parts of the audio very slightly so the voice becomes more clear. However, we don’t completely remove the natural volume changes of the voice.
Limiter
Finally, we use a limiter to further normalize the volume of the audio so we can push the levels without any clipping on our audio meter.
See my settings below:
My goal is to have the vocals peak at -6 dB on my audio meter. This will leave room for music and other effects so my mix is as close to 0 dB as possible without clipping. Because this is a mono track, setting the output level at 0 dB will result in -6 dB on the master meter.
Play your audio track, and boost the gain setting until you see the input meter reducing the audio. I like to set mine so that peak reductions are around 3-5 dB. I also change release to around 10 ms, lookahead to 4 ms, and mode to legacy with soft knee enabled. I’ve found these settings sound the best.
Final Result
And that’s it! Here are all the settings:
Check out the final before and after below:
What do you think of the difference? The voice is bright and clear, and sits well in the mix.
Adding these subtle tweaks can make a huge difference in the way the vocals sound in your final product. You protect your mix from going over zero, and everything comes together in a cohesive way!
