This is the complementary article to our “Audio Mixing in Premiere Pro.” The principles of audio mixing will transfer across platforms, but the tools are different.
Once you’ve completed your edit, you’re ready for the “finishing” steps of color grading and audio mixing. We know a lot of filmmakers enjoy coloring their projects because that’s when you can breathe life into your footage—especially when you’re using a flat color profile. But one aspect of post-production we hear filmmakers struggle with is the audio mix. The questions we’ve received about recording, mixing, and sound design could fill multiple episodes—we’re planning to create more in-depth audio tutorials in the future—but for this episode, we wanted to focus on a few tips and tricks we think all filmmakers should know to quickly improve their audio.
We’ve outlined four quick, and easy things you can do to improve your audio mix. If you’re working with professional equipment, and have all your settings dialed in, these effects will give your audio a final polish. Anything can happen on a shoot day, so while these tools may not be able to provide major audio recovery, they can at least help work around some of the most common issues you may encounter.
One quick note, this article and Ask the Editor episode is focusing on tools found in Final Cut Pro X. If you’re working in Adobe Premiere, feel free to check out Ask The Editor Episode 2 where we go over similar effects that are found in Premiere Pro.
There are a couple of things that you can do in order to set up the project library to make it easier to level out the audio:
The first things to get organized are the different audio roles. Labelling your audio roles will allow you to easily make adjustments to different audio types. The default audio roles are Dialogue, Effects and Music, but you can add new roles (as well as create sub-roles) to further define the different audio types. You can access the Audio Roles from the Index Tab, then click Roles on the far right. At the bottom of this window you’ll see “Edit Roles” which will allow you to create Sub-Roles and “Show Audio Lanes” which will group all of your audio files by type.
This comes in handy when leveling audio because you won’t have to scroll through the timeline to find similar audio types, they will all be lumped into their corresponding lane. You can add and adjust effects across similar audio types without having to copy+paste attributes. With audio lanes, you can also adjust the master volume on each Role once they are separated out by audio type.
If a video clip has multiple audio sources tied to it, you can set the proper channel from the Inspector window. This works if you have high quality audio tied to your camera, you can enable the high quality audio and disable the camera audio so you’re only working with the proper source. You can also switch between channels from here if you have multiple channels to work with.
Since we’ll be working with the audio levels, you’ll want to make sure you have the Audio Meters open so you can keep an eye on where they’re peaking. There are audio meters to the right of the timecode under the viewer, but they won’t help you determine the levels. If you double click on the meters they’ll open up and appear in the bottom right of your screen, and they’ll have the decibels visible. This will allow you to properly set your levels.
One final tip we have before you start sweetening your audio is to make sure you are making your adjustments with a good pair of headphones. Even if you have studio speakers in your editing space, you won’t be able to pick up on some of the fine details that you can hear with headphones. It is always good practice to play the audio back on the device you think your client will be watching the film on before finalizing it so you’re hearing what the viewer will hear.
The first thing we recommend doing to give your film a final polish, is EQ the audio. If the audio from your recording device feels a little harsh, or hollow, this effect can help smooth things out to make it feel more natural. There are a couple ways in FCPX to EQ your audio:
You can easily EQ any audio source from the Inspector window, under Audio Enhancements. The built in adjustments: Voice Enhance, Music Enhance, and Treble Reduce can quickly correct audio and give you a great starting point. It is beneficial to further tweak what is automatically set so that it works best for your audio.
To open up the effect window you can click the box to the right of Equalization. If you make adjustments here, it will set the EQ for the entire source file. If you add a different EQ effect to your audio, it will only adjust that clip. From here, you can start to adjust the different frequencies, and this will help bring some life into the audio. If you watched Ask the Editor EP 2, we talked a lot about this while going over the Parametric Equalizer, and this works similar to that effect. Bringing up the lower frequencies will bring some bass into the audio and give the source some depth. On the other hand, bringing up the higher frequencies will add clarity and give the audio a fuller sound.
Another way to adjust EQ, is with the effect Channel EQ which is also similar to the Parametric Equalizer in Premiere. If you want more information on how this type of graph effect works, you can check out our explanation of Parametric Equalizer in Episode 2.
If you’re looking for a quick way to adjust your music so it is not overpowering dialogue, you can use either of these EQ methods to notch certain frequencies.
First if you want to use the Channel EQ effect, apply the effect to both the dialogue and the music track (this is where setting up audio roles will come in handy). Open up the controls on each and arrange the two graphs side-by-side so you can view them at the same time. The male voice typically registers in the 80-160hz range, while the female voice typically registers in the 165-255hz range. Depending on who is speaking will determine the area of the graph that you want to focus on, make sure to turn on “Analyzer” in the bottom left of the controls in order to see where the waveforms are registering. Whatever appears to be the dominant frequency is the one that you’ll want to focus on.
Once you’ve found the dominant frequency in the dialogue, look at what is going on in that same range on the music track. On the graph for the music track, identify the high point of that range and then pull it down to about -6.0 db. You don’t want to pull it down too far as we’re just attempting to make room for the dialogue. You can then make further adjustments to the music track by keyframing the volume down. Adjusting the “Q” on the graph will widen or narrow the curve, which will determine how this affects the surrounding frequencies.
You can achieve a similar effect by using Equalization under Audio Enhancements in the Audio Inspection window. Open up the effect controls, the box to the right of Equalization, and if you’re using a song with lyrics you can open it up to the 31 bands and come over to the 12. This is roughly where the human voice registers, so pulling this one down will notch the frequency. Adjusting the points on either side will help achieve a smooth or tight curve.
What this process is doing is notching, or compressing down, those frequencies on the music track so they aren’t competing with the same frequencies in the dialogue track. Once you’ve done this, you can keyframe the volume of the music track, and in conjunction with these two processes, will allow your dialogue to have more clarity.
2. Noise Gate
Another helpful tool in FCPX is Noise Gate. If you’re getting a little bit of noise in your audio, you can use this effect to help cut out the frequencies that the noise is appearing in. This works really well for traffic noise, low hums like air conditioners, or hisses from microphone interference.
If you find a spot in your audio that has just the noise (this is why capturing a bit of room tone on the shoot day is always beneficial) you can see where the noise is registering and set the threshold to match this point. For example, if you have noise appearing at -20db, you can set the threshold of the effect at -20db. Reduction, refers to how much you want to reduce the noise by, but be aware that the frequencies are also in the speaker’s voice, so adjusting it for the noise will also start pulling it from the dialogue. Adjusting the Attack/Release will help smooth out any distortion you may be getting on the dialogue. This effect isn’t going to repair major noise issues, but it will help remove any distracting tones.
One tool every editor should be familiar with is Compressor. What the compress effect does is pretty much what it sounds like, it compresses the dynamic range of the source in order to bring your audio into the area you set. If you have a piece of audio where the speaker’s voice starts high but then they slowly start mumbling, you can apply this effect to bring up those low moments where they are speaking softer to better match the high points. When using this effect, make sure the Auto Gain is turned off, because if you’ve used other audio effects on this audio clip, the algorithm of this effect can mess up those effects.
We recommend setting your dialogue to a threshold of -3db, and the ratio seems to be best at 4:1. The Attack and Release can stay at the auto settings, but if you’re getting a bit of distortion, you can adjust these two points to tell the audio effect how quickly it should be turning on and off. If you notice that the Limiter is pulling out too much of the volume, you can adjust the gain from the Output Gain slider. This will allow the volume to come up, but still remain within the parameters set by the effect.
4. Adaptive Limiter
The final effect we recommend for FCPX audio workflows is Adaptive Limiter. If you are working with audio that is plugged into a DJ board or directly into a mic and you’re getting some peaking audio, you can use a limiter to set the threshold to a more desirable range.
Add the limiter to the clip, navigate to the Inspector, and click on the box icon to the right to open the effects panel. The presets can work in a pinch, but setting your own parameters will help you achieve the best sound. You’ll want to set your Out Ceiling to the maximum you want your audio levels to be. We typically set our dialogue to register between -6db and -3db, and our SFX/Natural Audio and Music when it is under dialogue around -18db and -21db. For the Release, we recommend setting it to between 300 and 500 ms, which is right around half a second, this is how fast the audio effect turns off so adjust it to what best fits your audio. If you’re getting any distortion from the effect, this will be the point to adjust to help correct that. The auto setting for Lookahead should work fine. Then set the Gain to where you want your lowest point of audio to come up to. If you look at the Reduction column you’re going to want to keep it around 1 to 3db. The Output point we set earlier is going to bring the audio ceiling down to that threshold, so any high points will not be able to go beyond that setting.
Like I said earlier, this isn’t going to help you achieve major noise reduction or repair anything that went wrong. If you have all your recorder settings dialed in, these effects will help sweeten the audio you already worked hard to record cleanly.
You work hard to craft a beautiful narrative, make sure everyone can hear it!