Audio quality is everything

I’ve originally put this simple audio editing guide together for internal use, since our marketing and design teams produce most of our videos in-house. I’m sharing it so others can learn from it too.

I’m gonna show you the few of the things that I think are most important.

I’m usually using OBS to record my own screen, or for online events with multiple people, we use a platform like Livestorm.

You can record yourself with a separate camera, audio feed on a mic, and screen footage as a third channel, and put it all together in OBS (putting the two or three screens next to each other) instead of syncing them all up manually.

(Then later, you can of course do a compound clip of the face part on top of the screen recording part.)

Optional video tracks you can have, each potentially coming with their own audio tracks.

You should have the computer’s audio and the mic’s audio on separate tracks, so you can edit them separately.

My recommendation:
After recording, take your mic’s audio into Audacity, and process the voice first before starting editing the video itself, so that you can already hear a nice, crisp, final version of the audio as you’re editing

It will also make it a bit easier to edit, because the audio track looks cleaner.

On the top, you can see the processed, final audio track and the raw version at the bottom.

I’ll not list all the specific settings in Audacity, as they are nothing too special, and you might wanna do something else based on your project, but I’ll give you the most important steps of the processing:

1. noise suppression (noise reduction)
2. noise gate
3. compression (legacy compressor)
4. low pass filter (for sibilance)
5. limiter (legacy limiter, to cut off any remaining spikes)

What’s great about Audacity is that it’s very visual, you can see how these changes make a difference intuitively.

The goal with these settings is to have it totally silent when no one’s talking, and when there’s talking, we want it between -5 and 0, regardless of what volume anyone’s talking at any given time.

That’s why you do the compression, because even when you talk really quiet or really loud, it ends up being pretty similar volume. Which is much nicer to listen to imo. As a viewer as I’m watching a video I don’t like when the volume drastically changes, and I have to turn my speakers up and down all the time.

The low pass filter helps with really sharp, high-pitch distortion of voices that some mics let through when you say an “S” sound for example.

So doing a lot of compression and some filtering and a basic limiter can make your audio much more consistent and that is a big quality of life upgrade for your viewers.

The video editing itself can happen in any old software, we’re gonna use DaVinci Resolve as an example because it’s free.

The best advice I can give you if you’re just starting out is to use whatever is free.

DaVinci Resolve has all the features you need to start editing YouTube videos.

So this is what the timeline looks like. You might change the colors of the clips to keep it organized.

Get Lőrinczy Márk’s stories in your inbox

Join Medium for free to get updates from this writer.

Remember me for faster sign in

Now, navigate to the very bottom of the interface and click the fourth “musical note” shaped icon, which opens the Fairlight page.

This is the audio mixing section of DaVinci.

I’m only using it for one thing currently, which is a little bit of a side chain compression on the computer audio.

I used to just set whatever audio is recorded from the computer really low, compared to the mic’s voice. Like you saw in Audacity, the voice of whoever is speaking is in the -5 to 0 range, and I used to set the computer recording to -20 to -15 range.

My reasonig was that if the footage is always really quiet, then you can always hear the voice of whoever is talking.

But the drawback is if there are people also talking in what you’re recording, then the audience will have to strain to hear them too, and then get jumpscared when you start talking again into the mic. So this is the same volume consistency problem again.

What you do in this case, is you do audio ducking or side chain compressor on just the computer audio.

When the mic’s audio goes above a certain volume, it lowers the computer audio volume automatically. Now you can have both of your audio recordings around -5, and whenever you speak, it will be audible.

Try to keep this very subtle, so that it’s not so obvious and that it just sounds seamless, and then people won’t even notice it.

You can make a new audio Bus, which is like a group of tracks, that you can rename as “sidechain” or anything you’d like, and put your mic’s voice and any of your sound effects in it.

The compressor on the first audio track will use this Bus as a reference to lower the volume whenever it’s needed.

To set that up, double-click on the audio track’s little graph that you want the automatic compression on.

This ‘Dynamics’ menu will pop up, which looks super complicated. And it is.

But you only need the central panel, turn on the Compressor, and in its bottom-right corner select the track or Bus it should be listening to as a reference (in this example, the one called Sidechain)

And if you press play on your video preview and open this menu, you’ll see on the graph live how much the compressor lowers the volume compared to the track it’s referencing, which is neat.

You can just copy the compressions settings from this screenshot or experiment to find the ideal settings for your own thing.

So this is it, it’s pretty easy and fun, and goes a long way in sounding better.

Professionally sounding audio is lowkey the most important part of editing a video, especially starting out as a new creator, and you got people talking in your videos.

It’s way easier to watch a video with good audio and shitty video than to watch a video with really high-quality video but shitty audio.

It sucks to listen to shitty audio, and you should put a lot of effort into recording and processing good quality audio.

Many people these days only consume videos in a mostly audio-only format, where they set down their phones and it just sort of plays in the background as they go on about their chores or jobs or whatever.

Sounding good, being nice to listen to should be your main differentiator compared to your peers

Of course I’m not trained in sound mixing or audio editing, these are just a couple of stuff I picked up along the way, so a pro audio person could probably pick this guide apart in five minutes and tell you to do everything oppositely, but I really like how the audio I edit sounds in the end.