Benny Mouthon (CAS), sound designer and re-recording mixer of CNN’s Anthony Bourdain: Parts Unknown, shares how he uses RX 6 Advanced on the show.
From a crowded kitchen in an Antarctic research camp to the busy streets of Hanoi, CNN’s show, Anthony Bourdain: Parts Unknown, shares stories from cultures all over the world. The show follows Bourdain as he explores new places and interviews locals, capturing authentic, first-person accounts over shared meals, no matter what lengths he and his crew have to go to record them.
Such a diverse range of environments and unpredictable surroundings create some unique challenges when it comes to editing and mixing audio. Luckily, they have the right man at the helm.
Benny Mouthon, sound designer and re-recording mixer for Parts Unknown, has been editing sound for Bourdain’s adventures since 2002, back when he was hosting A Cook’s Tour. Mouthon has been nominated for six Primetime Emmys, including Outstanding Sound Editing for Nonfiction Programming for Anthony Bourdain: Parts Unknown.
Without access to audio recorded by an on-set location sound engineer, Mouthon has relied on iZotope RX over the years to clean up audio yet still retain the natural sound of whatever environment Bourdain is in. In this interview, Mouthon shares how he uses RX 6 Advanced on the show.
“There’s nothing like getting interesting ambience from a special location, but you also have to understand that cameramen’s days are generally long and difficult; they will not have the time to go back to that unusual sounding market in Lima just so you can have a great, crew-dialogue free, stereo recording for your collection.” —Benny Mouthon, sound designer and re-recording mixer, Anthony Bourdain: Parts Unknown
How do you use Dialogue Isolate on Parts Unknown?
Except in special circumstances, Parts Unknown generally doesn’t use a location sound person.
The dialogue and natural sound tracks I get from the picture editor are recorded by the camera crew using lavalier mics (Tram TR-50’s with Lectrosonic transmitters) and camera mounted shotgun mics (Sanken CS-1).
During a meal, for example, the cameras will move around Tony and his guests, so I generally rely on the lavaliers as the foundation for the dialogue mix instead of the shotgun mics.
Anthony Bourdain has dinner with researchers and camp staff at the Lake Hoare research camp kitchen in the McMurdo Dry Valleys | Photo courtesy of CNN
Even so, some of the locations he visits can be rather loud: the restaurant might have walls made entirely of reflective surfaces like tiles, for example, or be packed with people who happen to be there on their lunch break or, my “favorite,” be completely open to the street.
In these tougher cases, I use Dialogue Isolate to control how much background noise I want to have as part of the mix in Tony and his guests’ lav tracks; I’ve gotten great results that have sometimes sounded “too quiet,” and I’ve had to ease off the processing!
It‘s simply amazing how, under the right conditions, it sounds extremely clean and artifact-free.
Which other RX 6 Advanced features do you use on Parts Unknown, and how do you use them?
I’m pretty sure there isn’t a single show I work on where I don’t use De-click, De-crackle and, since it came out in RX 4, Ambience Match, which for me is one of the most helpful plug-ins you’ve invented, especially since I never get room tone from the shoots.
Tony will often eat in locations that are not really favorable for clean audio: a beach in Cuba, by a Taco truck next to a highway in L.A., or on a sidewalk in Hanoi.
Once the scene is edited, the backgrounds might not match from sentence to sentence, and if I can’t remove it convincingly with EQ, Spectral De-noise, or Spectral Repair, I use Ambience Match to extract those problematic backgrounds so I can at least have a nice way to fade them out on the incoming clip.
Benny Mouthon using RX 6 Advanced on Parts Unknown | Photo by Lou Teti
Also, the lavalier mics are hidden and sometimes might scratch against clothing as people move about. De-rustle has been pretty good at saving those.
Have you used RX for sound design on Parts Unknown? If so, how?
On some shows Tony might exaggerate consumption of a particular food (like hot peppers) or alcohol and start to feel...uneasy. The editors might enhance those scenes by distorting the image or throwing it out of focus.
When it comes to audio, they layer sentences over each other to emphasize the fact that he’s losing control. In these situations I like to play around with a lot of reverb, delays, and panning the dialogue, but I also blend in over-the-top De-noised phrases (it gives it a “something’s not right” quality and helps give the impression things aren’t going well).
I also really like Iris; I used some of the “Wood” sounds toward the end of the show open, as the music starts to fade out. I’m also getting into using Stutter Edit, though that’s proven a bit harder to implement in Tony’s show.
You said, “I'm at the mercy of whatever audio the camera records.” Can you give me an example of audio you received from specific scenes that at first seemed untreatable but you ended up being able to fix?
There’s a scene in the Antarctica show where Tony talks to astrophysicist Dan Marrone in a camp that houses a huge radio telescope.
They had a very limited amount of time in that location before the plane had to leave, so they quickly unpacked the gear in the only corner that was free.
Unfortunately, this was set up close to some pretty loud helium compressors, which are used to cool down the computers that analyze the constant data from the telescope. The lavaliers were having a really hard time, and the shotguns were pretty much unusable.
I ended up applying a combination of Spectral De-noise and some Voice De-noise, which took a while since the ratio of noise to voice was not exactly working in my favor. You can tell that there are some artifacts, but in the end it was possible to understand them without being distracted by the excessive machine noise.
Whether it’s street noise or otherwise, I imagine that in some scenes you may want to leave audio “imperfections” in to make the scene sound natural. How do you make decisions to treat audio? Are there questions you always ask yourself, or a process you go through to determine what and how much to treat?
The show is first and foremost a travel show and, yes, I’m able to do amazing things with RX to rescue problematic audio, but I also have to respect the fact that what I’m given is as much a part of that location as what I’m seeing on the video screen.
For example, a hawker center (food court) in Singapore tends to be a loud place, and if I use Dialogue Isolate to separate Tony from the background, it will be cleaner but might sound too “empty.” Replacing what I’m missing with clean SFX ambiance is not always an answer, especially when dealing with exotic locations (you don’t want to use the wrong language!). If the location is loud, chances are Tony and his guest are talking loudly to each other, and that tends to sound strange when it’s too clean.
Basically, before I reach for plug-ins, I ask myself, “What is the least intrusive way to deal with this problem so the viewer understands what’s being said?”
On the other hand, the editorial team has been dealing with the footage for two to three months and know Tony’s dialogue by heart. It’s up to me to make sure someone who’s watching for the first time will get every word.
For other post production audio engineers who are not fortunate enough to be working with an on-location sound guy, what advice do you have?
Talk to your producers early and often! They are the ones that live with the audio the longest: they’re on the shoot, then sit next to the editor for three months, and finally, come to your screenings and hear what you were able to do or what could have been done differently on location to help the edit and mix be even better.
A good producer will try to find you before they leave and go over what they’re planning on doing; for example, “Tony’s going to see Youssou N’Dour play a couple of songs in a small restaurant in Dakkar.” Hopefully they can grab a feed off the board and place a portable stereo recorder discreetly on a table for crowd ambiance (you’ll be surprised how much a cheap Zoom recorder can add to a mix).
Also very important: chances are, the cameramen are listening with earbuds in one ear only, so they can hear what’s happening in the room. Since you will most likely use the lav tracks, they should make sure the one they have plugged in is the lavalier and not the shotgun mic.
If the battery on the lav stops working or the clothing shifts an hour into a full day and only records rustle, no one will notice.
There’s nothing like getting interesting ambience from a special location, but you also have to understand that their days are generally long and difficult; they will not have the time to go back to that unusual sounding market in Lima just so you can have a great, crew-dialogue free, stereo recording for your collection.
Most of the time, mono b-roll is all you get! If you’re lucky, you get enough to fill two tracks and pan them hard left+right so you at least get some believable location sound.
What have you learned about audio post production from working on Parts Unknown?
In these types of shows, imperfections in audio are something an audience is often willing to put up with; it’s understood that the sound of a location is as much a part of the story as Tony’s interactions with the locals.
Each destination is unique, and the sounds I’m given represent that, and sometimes it’s okay to let things be.