Did you know YouTube grades your video based on volume density in your videos? You know how commercials flip on and the volume blasts you away? Well this same thing can happen inside your video if you’re not careful. I’m going to show you how to avoid sharp peaks in audio (or dips) using Davinci Resolve. YouTube may penalize you not to mention your audience wanting to turn your videos off!
Getting started with Fairlight Audio Panel in Resolve
First a notice to the reader. I’m not intending to reinvent the wheel. There’s lots of GREAT tutorials out there already. For the most part this article will reference them.
It’s important to understand the basics of the Fairlight panel in Davinci Resolve. There you will need to edit your audio so it matches your overall production. Levels need to be taken into consideration, tracks cleaned up (less noise), and sound shaped (equalized / filtered) to give the right tonality. I found an excellent tutorial linked below going over all this:
Beginner’s Guide: Fairlight Audio Editing and Mixing in Davinci Resolve
Shaping your audio
Once you have a good idea of how the toolset works you’ll want to tweak effects / plugins to shape your audio.
Shaping the sound makes it much nicer than straight from the microphone. I watched this video and created my own equalizer presets. I now use the preset in Davinci Resolve for male voice. There’s a couple presets which are for a male voice I find work nicely for what I do.
Here is a screen grab from the equalizer on a vocal track in the Fairlight panel. I suppose if you had a lady in your production you’d want a second track to equalize her voice accordingly.
This video gets into mixing your audio tracks together. A common thing is to add music, even embellish the sound with folly (background noises). You don’t want to overwhelm your dialog though. Here’s where we get into sidechain and compressors. The audio sidechain (tracks you want to protect) gets fed into a compressor that pulls audio levels down on other tracks so they don’t overwhelm your dialog.
Dialog effects chain and compression
You’ll need to start with a clean foundation which is dialog in most video productions. The direction you head and how you tweak each effect will be highly dependent not only on the sound you want to achieve but also on type of equipment you are shooting on. Long story short, get your dialog where you want it to be. Then embellish with folly and last with music if desired.
My track A1 comes from the A-Roll camera’s built in mic / shotgun mic. Sometimes I use lav mics in which I then need to capture more folly / ambient sound with another audio recorder. I digress though, with a lav mic I can increase the voice isolation. Where as the shotgun may capture some of my ambient as well. How you layer your audio will take some careful though.
What seems to be consistent though is a filter chain as show in my screen capture (orange box for dialog). Compressor settings (dynamics) are highly dependent on your microphone setup as well as EQ.
I did observe though that my audio get lots of digital distortion if I incase the output level on my soft clipper to make up for the clipped high end.
Music compression
I created a number of videos with really high ratio compression on the music. This sounded terrible. The music volume harshly drops during dialog then spikes back up after.
I’m currently experimenting with a low ratio compressor and then pulling my music track down to like -18dB. I’ve also increased the hold time to 1.3 seconds as well as dampened the release. I found adding a knee in there (softening the edge by compressor) made the sidechain protection less noticeable.
Automatically add fades to your audio clips
I add fades to all my audio clips. This helps ensure you don’t have any harsh clips on transitioning from one audio clip to another. You can configure the fades you want and then clip “Apply Batch Fades” as seen above.
Adjusting audio mix for ideal Loudness Threshold (LUT)
At this point you’ll need to start tweaking all the audio pieces in Fairlight so you get a consistent audio level and or achieve your desired audio mix. This video shows how to utilize the tools to monitor Loudness Threshold measurements and tweak accordingly.
You will need to configure the tool correctly. Refer to screen capture above to get the Loudness history brought up. You’ll want to reset, and then start your Loudness meter (next to hamburger menu) before you playback your production from the start. Play it to the end and it’ll show you audio level tends as well as overall LUT you achieved in your edit.
Play your timeline all the way through and it’ll give you the waveforms shown below on Bus 1. This is where the real work begins. You’ll want to adjust this till the “Integrated” line hovers right around 14 and the overall is right around 14.
I’ve watched a few tutorials and they make this sound as easy as adding a limiter effect on Bus 1. It’s never as easy as they make it look. Previously I’d play back the timeline over and over again looking at “Short” reading. Then adjust clips slightly. I know the target is the Integrated but this takes from the short reading.
I just realized this waveform can be generated though, where you could display the Short Term and or Momentary wave on Bus 1 and try to bring up and down levels accordingly, saving you from playing the clip over and over again.
I also have to constantly revisit compressor on the dialog which seem to be the biggest culprit for spikes in audio levels.
I found it’s also easier to start with tracks under volume rather than clipping! My dialog starts at -6db, folly and ambient at -9db, music at -24db. Then you can use your limiter on Bus 1 to bring up your peak level to -1dB or whatever your target platform demands (e.g. YouTube).
Few additional tips to make dialog punch through music
This process can easily take as long as editing the video! No wonder they hire “sound engineers” for these larger productions…
Maybe the tool can replace them?
Automated Loudness Targeting in Davinci Resolve 19
Resolve 19 can attempt to automate audio normalization. This is done in the Render Page as shown below. I utilize custom presets so I update the preset accordingly and save it.
The automated tool can only do so much. It may even adversely affect your video? Use with caution, and definitely review your final render before you publish it! I am new to the automated tool in the render pane so still learning. Feel free to reach out to me with any questions you may have.
Results! Before and After…
I have only recently delved into audio editing using the Fusion page. Most of my videos only make use of the edit page with minor level adjustments to the clips. I first started fiddling with the Fusion page in Resolve 18 and quickly bogged my system down. Looked into purchasing the audio expansion card but then realize Resolve 19 optimized audio and there was no need for the card so I upgraded.
My first project in Davinci Resolve 19 with audio editing in Fusion page:
I used too much compression on the music track. You notice it when I’m talking and pause. If there’s music in the background it quickly peaks. I find these little details annoying, perhaps many people don’t even notice. I revised the music track compression in my latest video (will feature below when completed).
A project I did before I learned to do any audio editing:
The dialog and the music don’t seem to blend as well so I didn’t often dare layering the two. In hindsight, I do like the bare dialog w/o any music. However, when there is music it just feels like it’s apart from the dialog and the two aren’t meant to be together. It’s also interesting to note this was one of my first projects using Davinci Resolve after switching from Sony Vegas.
