In this part of Pro Audio, I’ll share the exact plugins (and settings) I use when editing and mixing episodes of a podcast series, Talking Too Loud, from the video hosting company Wistia. I would also like to provide a trail for those of you who want to use the share plugins that come with the DAW software, such as Pro Tools. This should also be easily translated into other software such as Logic, Reaper, Studio One and others.
How Wistia registers three people in three places
All episodes of Talking Too Loud have three participants – the host (Chris Savage), the producer (Sylvie Lubow) and an interview guest. As there are three people in three places, Wistia uses Zoom to facilitate the interaction between all the guests. Chris and Sylvie each use their Shure SM7B for a Zoom recorder, while the guest sound varies from episode to episode. All Zoom participants also record the streamed sound as a backup. (There are settings in Zoom so that each participant’s sound is recorded in individual files.)
Note: Zoom is mentioned twice above; There are actually two products that use the name Zoom. Zoom can be used to refer to the popular video conferencing platform and can also refer to the brand that offers portable recording devices.
• Zoom video conference
• Zoom recorders
To get started, here is the list of plugins I use when editing and mixing episodes of Talking Too Loud. I usually use Izotope plugins to handle the cleanup of the raw sound, and the Waves Renaissance series plugins for EQ and compression processing. I like the Renaissance series for this workflow because the plugins have simple interfaces, sound quite transparent and allow me to quickly dial in the settings.
Before I do a single edit, use EQ or use compression, I process all raw audio tracks offline with one or more of the Izotope plugins listed above. I generally use very gentle settings and then do some random sampling to ensure that the raw sound has not been adversely affected.
Background Noise reduction
After the raw sound is processed to remove serious plosives and clicks, Waves X-Noise is used to suppress background noise. Since everyone records from home, sources of unwanted noise can include air conditioning, refrigerators, computer fans or an elevated noise floor from microphone pressure. To use X-Noise, simply find a part of the audio track that does not contain speech and loop plays that choice. Then click the “Learn” button – this will create a noise profile for that track. Once X-Noise has been “trained”, you can only dial in threshold and reduction settings so that the background noise is suppressed while retaining the sound of the originals. Repeat this process for each track that has unwanted noise, and you should have much cleaner sound to edit and mix.
The EQ and dynamics chain
Now that the background noise has been suppressed, the rest of the processing chain is processing our smoothing and dynamics. The chain for each voice track is a straightforward combination:
- Renaissance Channel Strip ->
- Renaissance DeEsser ->
- Renaissance Compressor ->
- F6 Dynamic EQ ->
- Vocal Rider
Renaissance Channel Strip
The Renaissance Channel Strip offers EQ, gate / expansion and compression in a compact, easy-to-use interface. For podcast audio, I usually use EQ to use only a high-pass filter, usually somewhere between 80 and 100 Hz. This helps reduce rumbling or other low-frequency noise that can adversely affect comprehension.
Secondly, I make sure to deactivate the port part of the channel strip – I do not use ports, as I prefer to edit the silence manually for each side of the call.
Then apply a gentle compress. The goal here is to achieve some control for the track, but not flat dynamic range completely. I am generally looking for 2 or 3db gain reduction at this stage. Remember that we will use more compression with DeEsser, Renaissance Compressor and L3 later – we stack compressors in such a way that none of the individual compressors work too hard. This will ensure that we maintain a natural sound even when we take the dynamic range under control.
Check the settings below to get a good starting point. You will probably need to adjust the high-pass filter and the threshold settings for the compressor to suit your source material. Sometimes I want to do an extra EQ move in the middle range if something bothers me, although I usually wait until the sound hits the F6 Dynamic Equalizer (later in the chain) to do this.
Depending on the speaker and microphone, the source track may have unwanted weight, especially for “chh”, “shh” and “ace” sounds. A DeEsser is a great way to tame these disturbing sounds. Here are a few different settings I usually use on spoken tracks, depending on the speaker’s tendencies.
The goal here is to achieve some control over sharp “aces” or “chh” sounds without giving the speaker an artificial lisp. If you take the time to find the right balance, the rest of your podcast mix will come together relatively easily.
When our podcast sound hits our next plugin, the Renaissance Compressor, we’ve already used a high-pass filter, gentle compression, and treated hard sibilance with a DeEsser. Using the Renaissance Compressor will give extra control over the dynamic range, and since we are stacking more compressors, we can continue to use settings that are not too aggressive – which will help us maintain a more natural tone. Like the compressor in the duct strip that was used earlier, we aim for a reduction of approximately 3 dB.
This is what these settings look like, and keep in mind that you may need to adjust the threshold setting.
Waves F6 Dynamic EQ
Waves F6 can be used to make just about any EQ adjustment you may need at this stage of the process. It is an incredibly flexible and powerful plugin, with six fully adjustable bands along with dedicated low and high pass filters. I generally use F6 to treat the midrange as needed, and find it especially useful if I want to smooth out some hardness that may be present in the speaker’s voice.
F6 can be used as a conventional EQ to manipulate frequency bands of your choice, or more powerfully, as a dynamic EQ where you set a threshold and an area to control frequency bands only when they break this threshold. If you are not familiar with the concept of a dynamic EQ, you should consider it a compressor that only affects certain frequencies of your choice. Below is a picture of a setting I use for one of the podcast hosts. In this case, I have cut out only a hint of low and low center information using standard equalization, and have a dynamic eq configured (band 5 in purple) to tame some hardness around 2.7k. It really only works part of the time, when something in his voice gets a little peak – and even then, only a maximum of 3 dB in reduction.
Waves Vocal Rider
The latest plugin in my signal chain on podcast tracks is Vocal Rider, which automatically keeps dialogue levels stable and does so without the “staining” of compression. Although capable of more, it works well in 99% of podcast mix scenarios when you just cast it on a voice track and set a wide range. You can adjust the sensitivity and target to your liking, although I usually let them have the default settings. In rare cases where it does not work quite the way I want, I will actually print the automated father movements to an automation track and make adjustments to the father movements as needed. I tend to prefer the “slow” setting to avoid and sudden, big father movements, but I have also had success with “fast” mode.
Waves L3 Ultramaximizer
L3 is the very latest plugin used in a podcast I edit and mix. It is instantiated on the stereo mix bus and achieves two goals: more overall level and a mild amount of restriction. This is a place where you can transparently give even more control to the dynamic range of the podcast mix. It’s a simple plugin that just works – use your ears, and you will know if you have set the threshold too low. I like to set the ceiling at -0.1 to guarantee that no online streaming service will reject the sound for clipping.
Conclusion and options for stock plugs
Now that I’ve shared my voice processing chain for podcasts, you can experiment by replicating much of this functionality with the plugins you already own. The key to maintaining a smooth, neutral and balanced tone is to avoid aggressive compressor settings that color the sound. By stacking multiple compressors, you can gain control over voice recordings, yet still allow their natural dynamic range to come through in the final mix.
Although I’m not aware of any software noise reduction plugins that come with Pro Tools for free, there are some very useful options for EQ, compression and de-essing:
- Waves Renaissance Channel Strip -> Avid Channel Strip
- Waves Renaissance DeEsser -> Avid Dyn3 De-Esser
- Waves Renaissance Compressor -> Avid Dyn3 Compressor / Limiter
- Waves F6 Dynamic EQ -> Avid EQ3 7-band equalizer (not a dynamic EQ)
One last tip – the most important thing about putting out a podcast with professional sound will always come down to the quality of the original recording. If you take the time to make sure your recording environment is quiet and invest in a high quality microphone, you have already set yourself up to succeed. Wistia, for example, makes my mixing process easier than it could otherwise be, in part because they’ve invested in great podcast microphones (Shure SM7B) and individual recorders (Zoom H5) that capture pure, transparent sound.
The most important thing about posting a podcast with professional audio will always come down to the quality of the original recording. Click to Tweet