Updated: Dec 6, 2020
Podcast post-production is very important to gaining the trust of your audience. You don’t want them to tune in and constantly have to turn the volume up and down, or have such grading audio that they can’t listen to the entire thing.
I’ll be going over 3 main topics when it comes to how to properly and professionally produce your podcast in post, Organization and Restoration, Editing and Assembly, and finally Mixing/Mastering and Exporting. These topics will work no matter what DAW (Digital Audio Workstation) you are using to edit your podcast.
Organization and Restoration
First thing to do with your original audio is to make sure it’s well organized. If you ever need to go back and find a specific episode, or change an aspect of an episode, you want to be able to find it easily.
What I do is create a hierarchy of folders, first a folder for the entire show, then an episode folder, and inside that I have a folder for the original files, and the project for whatever DAW you are using. After I have everything where I need it I bring the files into the DAW and run whatever noise reduction process you have available.
A great way to accomplish this is using a ‘noise print’ function, I believe most DAWs have this capability. You find a section of audio where it is just room noise, capture a noise print of that moment, and the module will ‘learn’ what noises it needs to remove from the file.
After noise reduction, I apply other restoration techniques if needed, like mouth de-click, de-reverb, or de-breath. All of these aren’t available in every DAW but the purpose is to help your audio get to a base level of sounding good.
Editing and Assembly
The next step in post production is editing. What I mean by editing is cutting out anything that doesn’t add to the context of the episode. Any crutch words or stutters, or sentences that don’t lead anywhere. Anything that you don’t think helps move the conversation forward or isn’t necessary.
Editing is not an easy task, and is by far the thing that takes the longest when it comes to podcast post production, but if you want a clean, professional sounding podcast, it’s a necessity!
A few tips to get the best results, make sure you are not cutting off breaths! This is the most common mistake I run across with new editors, there is no easier way to tell that there has been an edit than if a breath is suddenly cut short.
The more reverb that is present in the room in which the host/guest records, the harder it’s going to be to cut words. If you cut right at the end of an “uh” the reverb “tail” is going on for several milliseconds after the word, so if you are cutting an “uh” you have to also make sure the reverb tail is not present at the start of the cut, otherwise it will sound like a bad edit!
A bad edit is one that can be heard.
Next comes assembly of the episode. This is where you insert the intro and the outro and make sure everything flows smoothly. This is what the final product is going to look like.
Mixing/Mastering and Exporting
Correctly mixing your audio will help the listener keep coming back week after week. It’s one of the most important things to get right and be consistent with. If your levels are all over the place, and your audio is fatiguing to listen to every week, your listenership will suffer!
A signal chain refers to everything from the source of audio, to the output of the final product. In this instance I will be referring to a signal chain as the order of effects used in mixing audio. This, of course, could be a MUCH bigger discussion, and there are entire forums and groups about how to best mix your podcast audio, but I’ll be going over what I do in the simplest way that I can.
First, I’ll talk about the effects applied to each individual track then I’ll move to what’s called a BUS track (I’ll explain what that means later) then we’ll go to the master track.
For each individual track or person on the podcast there is a specific signal chain that is applied. First is the EQ which is used to remove any extreme low end and high end frequencies, which are only adding distracting information into the human voice. This EQ is also used to remove any harsh resonant frequencies by narrowing each band and sweeping through the frequency range to find them.
After the EQ I use a de-esser, which is compressing a specific set of frequencies at the higher end (usually between 3kHz and 5kHz) to even out some of the ‘harshness’ of the voice.
After the de-esser comes a form of volume automation using a plug-in called ‘Vocal Rider’ from Waves. This is the only step that cannot be replicated using stock effects in your DAW. It is a specialized plug-in that automatically turns the volume up or down depending on the threshold you set, evening out the volume of each speaker.
Then I use a single band compressor, using only about 3-5dB of noise reduction. So pretty light compression. This is to even things out even more, bringing the quieter parts of the audio up and the louder parts down.
After the compressor comes another EQ. This EQ is used to shape the sound a little more to what sounds good to your ears. This step isn’t necessary, but I like to sweeten the frequencies that sound good and slightly reduce the ones that don’t. I have recently started using ‘Soothe 2’ from Oek Sounds, which further removes harsh resonances from your audio automatically.
I then route each vocal track (any track with someone speaking) to a BUS track. This allows you to apply effects to multiple tracks at once, helping to glue together the podcast as a whole.
On this BUS track I apply another EQ which very slightly further shapes the sound, as to help the podcast sound more uniform. I also add another compressor here with about 3dB of gain reduction, accomplishing the same thing as the EQ.
Finally on the master track I am only applying two effects, a hard limiter with the threshold set at -0.1dB, to ensure that no audio will ever reach 0dB, which would cause distortion, and sound terrible. I also add a loudness meter in order to track the perceived loudness in real time. Which brings me to my final point in the mixing/mastering stage. Your final product should almost always be at -16 LUFS when exporting a stereo file, and -19 LUFS when exporting a mono file.
LUFS stands for Loudness Unit relative to Full Scale, and it measures how loud we perceive audio to be long term. It’s one of the most important factors to be measuring when producing your podcast. You don’t want one track to be significantly louder or quieter than the other, so make sure you are mixing them to a similar loudness level.
Exporting your audio is simple enough, most hosting platforms have a maximum file size that necessitates an mp3 format. For the highest quality mp3 you can export at 48kHz sample rate, 32 bit depth, and 192Kbps. That will give you the smallest file size with the most useful information still intact.
I couldn’t cover everything here, but I believe this will give you a major advantage on the other podcasts being produced out there today! Remember that post production will definitely help create a better end product, but it’s much more important to get the recording done right in the first place! Only so much can be done with bad audio! Go out there and make a great podcast!