I recently had a talented impressionist artist grace my studio to record a number of short scripts he needed for a showreel. This was a fun way of spending an evening but I had a lot of serious work to do in the mix to get these recordings to sound their best.
In this article I walk you through this entire session covering how approached the recording through to how I chose to process and mix the scripts to give you an idea of what to expect if you find yourself producing a dialogue artist who performs with an extremely large dynamic range.
The Session Plan
The artist prepared around 30 short scripts that we needed to record in the session. I’ve seen him perform many times in public. I know that he performs best when he’s in his own little bubble so I decided early on not to distract him or make him feel under the microscope by punching him in and out of record. I set the microphone up in an appropriate space in the studio, a space that gave him plenty of room to move around as well as being a good sounding space for his voice. I also gave him plenty of headroom on the mic pre to enable him to really open up his dynamic range without clipping the recording. When we had a sound we liked at the microphone we set Pro Tools to record and left it running until he finished performing all of his scripts in the knowledge that I would cut up and organise the takes later in the session.
Out of my microphone collection I reached for my Townsend Sphere L22 microphone as this system provides a range of virtual mic models that can be toggled between after the recording. It’s also a fine sounding mic. A couple of days before our session Townsend released an AAX DSP version of the Townsend plug-in:
These days it’s fairly rare for plug-in developers to release AAX DSP plug-ins for the Pro Tools HDX system, I should know as I use HDX in my studio. This release was indeed a very nice surprise. I used the AAX DSP version of the Townsend plug-in for the first time in this session, I can happily report that it worked a dream enabling me to monitor each of the mic models in recording at near-zero latency.
In this free video tutorial I use two of the scripts we captured that day to demonstrate how I processed the recordings.
In this video you may wonder why the audio waveform of the dialog recording appears to look like an uneven stereo recording. The Townsend Sphere L22 microphone we chose to record with requires a stereo track to record to as the microphone’s from a rear capsules need to be captured in order for the plug-in to work its magic. What would be the left channel in a stereo track is the front capsule (hence why the waveform is larger), the right being the rear of the microphone. The Townsend plug-in then sums the stereo track to mono.
This free tutorial is in no way the definitive way to mix dialog tracks, instead it represents one way you can mix a dialog recording that you can try for yourself or adapt the next time you find yourself tracking and mixing a spoken word performance.
Organising The Audio Clips And Trimming The Fat
After we finished recording the scripts I reached for the trusty Smart tool in Pro Tools to cut up the full length recording into short audio clips.
I also trimmed away any noise between the lines within the scripts such as my computer fan whirring away in the background and of course the occasional laugh from me. I shaped the breaths with fades and silenced any long pauses as there was some, if only slightly audible, pre amp hiss that I felt need to be tucked away.
Reducing The Hiss
To reduce the preamp hiss further in the dialog recording I opted to use iZotope’s Voice De-noise plug-in. I took a snippet of audio that had no speech in it which represented the overall hiss I wanted to reduce. I set this to playback on loop, set Voice De-noise to learn and reduced the hiss using the threshold and reduction faders. The results of iZotope’s Voice De-noise are always very clean and very transparent.
Adjusting The Microphone After Recording
I was toggling between different microphone models in the Townsend Labs plug-in while we were moving the microphone around the room. At the time I wasn’t too sure which mic model was appropriate, luckily I didn’t need to commit to any of them in the tracking stage as all the settings can be changed after the fact in post. I did end up using the mic model I chose in the tracking stage as I felt it was the most flattering for his vocal tone. I did however exaggerate the proximity effect in the plug-in as this gave a nice low end quality to his voice.
Changing the polar patterns, adjusting filters, axis and proximity are all things to consider and commit to in traditional microphone workflows, with Townsend these can all be adjusted later in the mix after recording which makes this a very useful an unique system indeed.
Reducing Mouth Clicks And Pops
Mouth artefacts are fairly common in dialog recordings, depending on the artist of course. Our recordings captured a fair few short mouth splatters but this didn’t faze me as iZotope’s Mouth De-Click plug-in works wonders at removing these right out of the box.
Addressing Thumping Plosives
Plosives sound like low end thumps and can distract the ear from the narrative in dialog recordings. These can be reduced with high pass filters though you may affect the overall tone of a voice working this way. You can of course use multi-band EQ or multi-band compression to reduce plosives, but why not use a plug-in developed specifically for the job of addressing plosives? Accusonus ERA 4 Plosive Remover works wonders. Like iZotope Mouth De-Click you really don’t need to adjust anything expect how much of the process you want to apply by use of the main central orange dial.
Squeezing The Dynamic Range Without Pumping The Performance
Many of us on The Production Expert Team love Avid’s Pro Compressor. It does a great job of really squeezing the dynamic range of just about anything you through at it with minimal pumping artefacts, which made it the perfect compressor to reach for in this session.
Heavy compression can exaggerate sibilance in a dialog recording. A good way to reduce the harsh sounds of sss or shhh in your dialog recording is to use a De-Esser plug-in after the compressor. Accusonus’s ERA De-Esser Pro was my de-esser of choice in our session as, like the Townsend Plug-in, it too was also recently released so I decided to take it for a test drive straight away.
It’s as simple to use as its ERA single dial sibling, made better with the additional focus and shaping tools which do a great job of helping us to hone in on troublesome frequencies. I couldn’t fault De-Esser Pro in the session. Only a handful of the recordings needed some degree of de-essing. Top tip: Try not to over process sibilance as heavy de-essing can ruin the presentation and presence of a voice.
Final Touches Of Tone Shaping
We did boost some of the low end energy by using the proximity control in the Townsend Labs plug-in, apart from that the overall tone of the voice sounded spot on to our ears down the microphone and in the mix. I chose to filter away the extreme highs and lows as these ranges didn’t seem to add anything to the overall presentation of the mix.
As we stated earlier in this article, this free tutorial is in no way the definitive way to mix dialog tracks. We would love to discover how you process your dialog tracks. What plug-in chains do you use?