Joker: Folie à Deux – Dean A. Zupancic and Tom Ozanich On Crafting a Killer Mix

If you missed Director Todd Phillips’s Joker: Folie à Deux in theaters, you missed hearing a killer mix by re-recording mixers Dean A. Zupancic and Tom Ozanich (at Warner Bros. Post Production Creative Services in Burbank) in all the glory of a Dolby Experience or IMAX theater. The mix team delivered a singular cinematic experience that proves they’re at the top of their game.

Joker: Folie à Deux is available now via VOD, and will be streaming on Max on December 27.

Zupancic and Ozanich once again worked with sound team members Jason Ruder (executive music producer/supervising music editor), Steven Morrow (production sound mixer), and Kira Roessler (supervising ADR and dialogue editor), who all worked on A Star is Born and Joker. They know what’s needed to capture the magic of the performances on set and carry that through to the audience in their seats. The team welcomed Erik Aadahl and Ethan Van der Ryn of E2 as supervising sound editors/sound designers. They designed the intense world of Arthur’s reality – the world in which the dialogue and vocals live. Zupancic and Ozanich had the best material from editorial and had time to experiment and perfect ideas on the stage. The result is an incredible film mix that is captivating and bold.

Here, Zupancic and Ozanich break down their approach to the mix on the opening hallway scene, playing with perspective, working with dynamics in terms of volume, surround field, and frequency range, building the songs using live vocals, music stems, foley, effects and backgrounds to make them live in the world on screen, and much more.

Joker: Folie À Deux | Official Trailer

The film opens in a prison hallway. There’s such a wide sound field that’s filled with trauma and distress, which is seemingly intensified by the whistling guard walking down the hallway. The specificity and movement in the sounds are incredible. You know right away this is going to be an amazing-sounding movie. Can you talk about your use of panning and reverb to establish this location?

Re-recording mixer Tom Ozanich

Tom Ozanich (TO): That’s exactly what we were going for.

Dean A. Zupancic (DZ): We know it’s important to open big and use the surrounds. We needed that contrast from the mono cartoon that precedes the prison scene.

TO: Technically, it’s not fully mono. It’s more like a wide mono or a very narrow stereo. It’s a choice we made because we wanted to play up that contrast. That was something we wanted to play with a lot in the movie, in general. The point of view is always changing. There are all these minor shifts and subtle things, but that transition from the cartoon to the prison was one where we wanted to be big. We wanted to inform the audience – right off the bat – that we’re going to try something big here.

Re-recording mixer Dean A. Zupancic

We wanted to find how far we could push this three-dimensional space and make it feel very believable that you’re in the hallway with them. We pressed the limits of how accurately we could get the detail of the space, the reverb, and the positioning of all the different sounds and people calling out. It was a lot of fun to try to dial that in.

DZ: It was important for us to establish the space. We see the two guards walk down the hall to open those doors, and we really worked on having those footsteps echo down the hall so you felt the reverb where we were sitting, but the footsteps got more distant. We worked on that quite a bit to get that three-dimensional feeling. We love the open. All of reel one was just a really fun reel, sonically.

Were you using discrete reverb processing for each channel, for each voice we hear in the surrounds? Or, how did you keep the scene from being messy and watery with reverb?

TO: That is something I’m always very cautious of as reverb oftentimes creates these muddy mixes if you let it get out of hand. I’m always trying to figure out how I can sell the space with the least density and tail of reverb so we feel the distance or we feel the space but the reverb isn’t lingering so long that it sticks around and muddies up the next thing.

I’m always trying to figure out how I can sell the space with the least density and tail of reverb so we feel the distance…but the reverb isn’t lingering…

Going into this film, we knew what to expect because of our working relationships with Todd Phillips and this sound crew having done the first movie together. Even though Erik Aadahl and Ethan Van der Ryn were new to the crew, we’ve done a bunch of stuff with them and have similar sensibilities. We got the script and we were thinking about ideas before they even shot the movie. That allowed us to talk about some plans and ideas we wanted to be able to experiment with. So much of the time we’re thrown into situations where we don’t have time to experiment. We just have to do what we already know will work and then move on to the next thing.

For this film, we worked with our post-supervisor Lisa Dennis and the whole team to build a structure that was a bit unique in regards to mixing this movie. It was planned to have time to experiment and play with these things in the mix.

Oftentimes, I’m using a couple of different reverbs or delays at once, blending the processing to get the space that I want.

One of the things that happened technically was that Pro Tools came out with the ability to have up to 9.1.6 buses. I was waiting for that to happen. As soon as that happened, I said, we’re going to use that. There was some concern about how that would work with having so many stems, that it may become technically too cumbersome. But working with Warner Brothers’ engineering team, we figured out a way to do that and solve some of the latency problems that get introduced with that. We had two recorders. It was a big setup but it allowed us to use 9.1.6 reverbs and delays. Oftentimes, I’m using a couple of different reverbs or delays at once, blending the processing to get the space that I want. That can allow me to feed voices that are further down the hall into a deeper reverb versus voices closer to camera feeding a reverb that’s more of an early reflection. I can do this in different amounts so that it can build the layers of that three-dimensional space.

Right off the bat, we knew we were going to push the system to its breaking point and see what we could get out of it.

Some sounds in a reverberant space are going to trigger the environment more, such as transient sounds, sharp sounds, and whistling. Those sounds really bounce around a space. I love how you handled the whistling in the hallway. It’s not just bouncing off the walls but it’s also moving through the space. You can hear the changes in the panning as it’s moving, and the changes in the reverb as it gets closer to camera…

TO: Again, I wanted to sell that movement, to make it feel like he’s actually way down the hall and off to the right. He’s so far away that it’s a wash of reverb at first and you can’t quite tell. As he gets closer, it gets more and more specific. Then he comes by and there he is. You really feel like he popped in. Of course, the recording of the whistling is stationary. We achieve that movement through manipulation in the mix.

…the recording of the whistling is stationary. We achieve that movement through manipulation in the mix.

DZ: It’s not just reverb. You’re manipulating EQ as well, which is important.

TO: The whistling, in particular, is tricky for EQ because whistling is a very narrow frequency band. You’re only working with a narrow range of frequencies so the EQ is very touchy. If someone is talking you have a wider band of freqenices to work with. You can roll a bunch off of the top. But if I did that with the whistling, it’s just gone. So, you have to play with different things – some of it is level, and some of it is EQ – on a case-by-case basis.

The hallway scene sets high expectations for sound in the film. You put so much effort, love, and care into that hallway sequence. For a regular moviegoer (or a non-sound person), that scene is just setting up this place and these characters. But as a sound person, I was blown away. If so much love was put into a hallway scene, what is the rest of this movie going to be like? It made me excited to hear the rest of the film…

DZ: Thank you for getting it because it was important to grab the audience. We wanted to establish this frightening place with these happy/sadistic guards. The contrast is what we’re trying to sell through the opening mix.

Some of what you do is just to achieve a feeling…You’re trying to give them a unique and vibrant experience that they’ve never heard before

TO: I’m very excited that you recognize that and appreciate that because I love it. I wondered if anybody was even going to notice this.

Some of what you do is just to achieve a feeling. You understand more technically what’s happening, whereas somebody who’s not into sound is going to the movie just to have an experience. You’re trying to give them a unique and vibrant experience that they’ve never heard before, even if they can’t explain it.

Typically when we talk about dynamics, we’re talking about the difference between the quietest and loudest moments in a mix. We don’t often consider the difference between the narrowest and widest uses of the surround field. When it came to the songs, you showcased the use of the surround field in a unique way. Many of them start with a dialogue line that’s sung, and then the orchestra builds out into the surrounds. The song isn’t just expanding musically, it’s also expanding physically into the surround field. That’s amazing…

TO: Different pieces of music work differently. A great example would be the Frank Sinatra song that comes in as Arthur walks out of the prison yard. The song starts and it’s a typical needle-drop start – what you’d typically hear in a movie. But it’s weird because the guard, Jackie, sings a couple of lines from the lyrics. So you wonder, is he hearing this? That blurring of what’s real and isn’t real is always a factor in the Joker movies.

The span of the dynamics is the width and breadth of your canvas.

The song starts in this expected way but it slowly builds, evolves, and changes into something totally different. That was a construction that Jason Ruder (executive music producer/supervising music editor) did working with composer Hildur Guðnadóttir to build something there and adding elements from her score. Then we did some deconstruction processing on the Sinatra track to pull the vocals out and separate them from the music. Playing with that in the beginning was a lot of fun. There are so many various flavors and different styles in the music. It doesn’t do just one thing.

One thing I would say about dynamics is that it applies to a lot more than volume. It’s not just the quiet versus the loud. It’s the narrow versus the wide. It’s the low frequency versus the high frequency. The span of the dynamics is the width and breadth of your canvas. You try to push those boundaries out as far as you can (within reason) and utilize all of it.

Another great example of playing with the sound field is during Arthur’s interview with Paddy. We hear the TV interview as mono through the camera and when we’re in the jail space with them, we hear the sounds of jail reverberating in the halls. Paddy and Arthur’s voices are bouncing off the walls in the little cell. You feel the space they’re in, especially coming out of the interview as seen on TV…

TO: There are a lot of ambient shifts. In addition to the reverbs for their voices resonating in the room, we also have far-off moans and ambient sounds that make the space seem bigger beyond what we see on screen, beyond the room they’re in.

DZ: The room ambience narrows and pops every time we go into the camera and then come out of the camera. The room is also narrowing and becoming part of that camera viewpoint.

The movie gave us a lot of latitude to play space and dimension. We just went with it.

Even when his lawyer Marianne and the guard Jackie are watching the interview, the room they’re in is a separate room with different ambience and different reverb.

The movie gave us a lot of latitude to play space and dimension. We just went with it. We had time to experiment. When sound guys on a mixing stage get excited numerous times, we feel like we’re doing something special because it’s not the usual ‘get it down/do the best you can.’

This was a great experience. We have to give Erik and Ethan a huge amount of credit for the sound effects and sound design. What they did was extraordinary. Their work got the picture editor Jeff Groth and Todd excited, too. Early on, they would send ideas that they were having – stuff they talked to us about – and the picture department would expand on that idea and throw it back to Erik and Ethan, who would then expand on that idea. It was a process of growth. The great thing was that when Todd came to the stage to listen to a reel or a playback, he wasn’t surprised. He was already on board with the direction we were going sonically, which helped us experiment more. We weren’t afraid that he was going to kill an idea.

we had multiple attempts at questioning the scene. Is that truly the best version of that? Are there any other interesting ideas we can impart here?

TO: We had this unique experience of trying an idea, getting to a point where we were all happy with it, and then revisiting it later and trying something new – adding a sound to this section or making a change here. If it didn’t work, we’d go back to what it was. If it was a good change, that might give us another idea. Normally, you would have just been happy with it and moved on. But in this case, we had multiple attempts at questioning the scene. Is that truly the best version of that? Are there any other interesting ideas we can impart here?

Speaking of having great material on the dub stage, during Arthur’s trial, it’s a media circus outside the courthouse. There’s a whole crowd of fans and reporters out there, and sometimes during the trial, you can hear their presence in the courtroom. If you don’t have that material on the stage, you can’t put that in the mix…

DZ: It’s great that you pointed that out because that was a late addition to the mix. We were always trying to make the outside alive, but not distracting. So you’ll hear an occasional siren-by or traffic. But it felt like the courtroom was off the busy street and on a side street.

…it wasn’t something we had planned. It was an idea that evolved through the process of trying to make it better.

So one day Erik came to the stage and said he had an idea last night. He wanted to have a pro-Arthur crowd outside the courthouse. So we messed around with that, using different EQ and putting some slap on it. It was pretty cool. It worked great. Then Todd came in and we showed him the idea and he liked it. But it wasn’t something we had planned. It was an idea that evolved through the process of trying to make it better.

TO: What’s so awesome about it is that the crowd outside reacts to what’s happening inside the courtroom but it’s a delayed reaction, as if they’re getting a news feed. When they find Arthur guilty, there’s the courtroom reaction and there’s a beat before we hear the people outside react to the verdict.

Arthur gives an impassioned speech to the jury. He’s not being the Joker here. “It’s just me,” he says. He’s using a handheld mic on camera that captures all the emotion, feeling, and imperfections like breaths and pops in his ‘performance.’ Can you talk about your dialogue work here?

TO: The production audio wasn’t different whether he was on the mic or off the mic. We did that by EQing it differently, changing how loud it played, how it fed into the reverbs, and so on to make it seem like it’s that handheld mic that we’re hearing.

That was something Jeff (picture editor) was super into as well. We spent some time trying to be specific about how the mic sounded based on what we were seeing, whether it was out of frame but still up close here, or a little further away there. It’s really motivated by what he’s saying. You want to feel this closeness and proximity to him at that point. That adds to the emotional depth of what actor Joaquin Phoenix is doing.

You want to feel this closeness and proximity to him at that point. That adds to the emotional depth of what actor Joaquin Phoenix is doing.

The witnesses on the stand are in front of a microphone, the judge has a microphone, and there’s one on the desk that Arthur talks into a few times. Early on, I tried to establish that those things are there without being heavy-handed about it in the mix. So when somebody leans into one of those mics, you get a little more of a feel that there are cheap ’70s speakers in the back of the room that the courtroom crowd can hear. You hear a little bit of room and slap on that futzed version of what the people are saying. That helped to establish and build a world in the courtroom so when we get to Arthur’s big confession, we can go even further with that and have a bigger push to proximity on the mic with him.

DZ: One of my favorite mic-handling moments is during the “Joker” song. I love the treatment that Tom did with the closeness as we come around. That was great.

TO: I love that, too, but it’s a total cheat because Joker is holding the mic the whole time. In theory, we wouldn’t be hearing this proximity, but it just felt right to hear it. That’s probably a moment where AI would choose to do something different because it doesn’t necessarily make logical sense, but it felt right. As the camera comes right by him, we get that close proximity. I wanted to feel the difference between when he’s singing with the mic versus when he throws the mic and is just singing in the room without it.

In theory, we wouldn’t be hearing this proximity, but it just felt right to hear it.

There are so many little things like that throughout the whole movie (not necessarily related to a microphone) that reinforce proximity and small panning moves that move people off of center so they feel a little less focused, and in some cases, a little less important. There are times when Arthur is sitting there just listening to the trial go on. He’s maybe drifting off a little bit, so we let the conversation drift off a little bit. There isn’t really something else to listen to. It’s not like he’s thinking about some other conversation or some other thing that we’re focusing in on. The trial just drifts off a little bit and you feel a little bit of a distance. Even though you’re still tracking the trial and what’s being said, you have a sense that he’s not paying that much attention to it. He goes along with the trial as if it were a show. He agrees with statements and makes arguments that aren’t helping his case. He just loves the entertainment value of it.

After the car bomb explosion outside the courthouse, we experience Joker’s POV with tinnitus. All the rubble and debris, his breath and movement sounds, and the people talking outside on the street are rolled off with full frequency score playing over top. Can you talk about your mix choices for this sequence?

DZ: The tinnitus idea came from the cutting room and it was a general idea in the temp track – just an idea that they wanted to go in that direction.

We got together with Ethan and Erik, watched the rough cut, and decided we needed to do something different so the tinnitus scene in the movie didn’t sound like every other one.

…we needed to do something different so the tinnitus scene in the movie didn’t sound like every other one.

Erik and Ethan did some experiments and we were leaning in the direction of hearing Joker’s movements muffled, hearing low-frequency sounds as he crawled across the floor and through the rubble. It grew from an idea of tinnitus to that.

As for the score, that was the only way to play it because this was a turning point for Joker.

TO: There is a little bit of that typical high-frequency tinnitus sound following the explosion as Arthur is lying on the ground. But, we didn’t want to play that for too long because it’s not pleasant to listen to.

Also, we didn’t want it to just sound like it was simply muffled. It needed to have some character to it. We wanted to have the kinetic sounds of him moving around, as though he’s hearing through his bones – almost like a contact mic versus a muffled mic.

I wanted the score to slowly emerge out of this sea of murkiness that he’s hearing …

As he gets up, the score starts to come in. I wanted the score to slowly emerge out of this sea of murkiness that he’s hearing – let the score start to bubble up out of that and grow into this overwhelming, almost victorious thing since he’s gotten free. Then, it quickly throws you for a loop and shifts and we did some fun Atmos panning of elements as he’s walking towards the car with another person who’s dressed as Joker. There’s this slow evolution of him getting his hearing back.

DZ: There’s a sound effect for Arthur getting his hearing back. It’s very subtle, but you hear this little sound of his ears popping/opening up. It’s a great moment because after he pops his ears, you hear the Billy Joel song in full frequency on the radio and the guys arguing about which way to go.

It’s another example of how we were using dynamics without hitting you over the head with dynamics.

TO: In any mix we do, we always try to work with dynamics in a way that’s interesting and pleasurable to listen to.

Can you talk about your approach to the songs in the film? Most of the songs start a capella and then slowly build out into full orchestration. Some of the singing is so up close – like Arthur’s rendition of “If You Go Away” into the phone during his call to Harley, and Harley singing in the jailhouse visitor’s cubical the “(They Long to Be) Close to You” song. Were these on-set performances?

TO: All the vocals are happening on set.

The songs in this movie were basically live on set. They each had their own piano player in an isolation booth playing along with them but following the actor’s tempo. The actors were setting the feel. Afterward, Jason [Ruder] would take these raw live performances – without a click track – and structure and build the songs to what was performed on set. It was a totally backward way of doing that but very effective in making it sound believable because in all reality they were singing on set.

Jason [Ruder] would take these raw live performances – without a click track – and structure and build the songs to what was performed on set.

We did a lot of things on this movie that we don’t normally do. The performances worked a bit like dialogue. Essentially, Jason and Todd would choose takes from the dailies that they liked and our dialogue supervisor Kira Roessler would get that production dialogue/vocals and build it as if it were dialogue, which is different than comping vocals for a song. Then, she would give that edit to me and I’d clean it up and make it feel more natural – like I’d typically do for dialogue. Then, we’d hand that back to the music department who would tweak things as they needed to and then they’d hand it back to me. I would then do the next level of treatment and processing to it.

…he’s very intimate, low level and so consequently, it’s a pretty noisy track.

There was a lot of back and forth that would normally not happen. There’d be problems, maybe, from the production tracks that we’re stuck with because the music department did what they did and now that’s baked in and we can’t change it. We tried to avoid that but a lot of the production tracks were super noisy. For example, in the song you pointed out – of Arthur singing on the phone – he’s very intimate, low level and so consequently, it’s a pretty noisy track. There were a few other spots as well that I thought we wouldn’t be able to use because it was really problematic. It took multiple passes to get it the best that I could, and then we’d all listened to it. I’d ask Jason if there was any way we could get a couple of words from a different take that might be cleaner. In some cases, we did that. And in some cases, we were able to take it a little bit further. Again, I had a couple of different stabs at it to be able to get it as good as I could and then walk away from it and come back the next day with fresh ears and work on aspects of it to try to get it better.

The music stems were constructed after the vocals. The songs were delivered as 5.1 stems that I could break into parts and use in the Atmos surround field.

The score was all quad-based stems as we had done for the first Joker film. It is a very effective way of using those stems for Atmos.

There’s a scene inside the courtroom when we’re in Arthur’s headspace as Sophie Dumond gives her testimony. This transitions into Arthur imagining he’s on stage in the spotlight singing “The Joker” song. He smashes the stool and throws the mic, shoots the TV cameras, and beats up the judge. It’s a very energetic scene with singing…

TO: For the most part, the gunshot, the beatings, and smashing objects all happen in between a line of dialogue/vocals. It was pretty much all effects with maybe one or two production effects.

We’re trying to heighten his fantasy with the shock of the gunshot and even his last hit on the judge before he shoots himself. That’s pretty visceral, pretty brutal – all very intended.

DZ: We’re playing those effects brutally because Joker is brutal. We’re trying to heighten his fantasy with the shock of the gunshot and even his last hit on the judge before he shoots himself. That’s pretty visceral, pretty brutal – all very intended. You hear laughter in the courtroom but also shock and horror, so we’re amping that up to show that this guy is off his rocker. That’s one of our favorite scenes.

TO: It’s a good example of dynamics, too. One of the gunshots is shooting into the camera we’re looking through so there’s glass that breaks in the back of the room. That was something we were playing with and Erik wanted to add other details, other elements. We weren’t being shy about those things.

DZ: That gunshot is pretty shocking, and the car bomb explosion was too. If you listen to the car bomb, there’s a pre-explosion sound. That’s how it would work in reality. You have the pre-explosion before the big explosion, like “crack..BOOM”.

What about the song where Arthur and Harley are dancing in the rain? This starts in the music room that catches on fire and then they’re outside in the prison parking lot, dancing in the rain and singing. This must have been challenging!

TO: Production sound mixer Steven Morrow did a fantastic job of recording in challenging conditions. You have rain machines, fires happening, people running around, and all of this stuff happening. You’re going to pick up a lot of these other sounds and they don’t necessarily all sync with what you’re seeing some of the time. A lot of that has to be removed or drastically minimized. There were times I was just sitting there thinking, “I don’t know how we’re going to be able to use this.”

Dean is a master at putting in just the right amount of foley, backgrounds, and effects…so the world feels like it’s all from that production recording.

But, you just keep working on it and not giving up. The performances are so great that you feel like you’ve got to figure out a way to make them work. You figure out a way to clean it up without sacrificing the actual quality of their voices. They still need to sound like themselves. Sometimes you do one thing and that creates a problem, like making something more sibilant. So you have to attack that problem. You’re balancing how far to go with any one of these processes to find a middle ground.

I rely on Dean a lot to help wrap that stuff in the world so that if you were to listen to only that vocal track or that dialogue track, you’d be like, “What is all that weird stuff going on?” Dean is a master at putting in just the right amount of foley, backgrounds, and effects to sit right in there with the dialogue and the vocals so the world feels like it’s all from that production recording. You’re not aware of any of the problems but those problems still exist in there to some degree or another. That’s why I rely on the master.

DZ: Ditto. It works.

It also helps that we’ve all worked together before. We’ve worked with Steve [Morrow] several times since A Star is Born. He knows how we work. It was a good bunch. This is one of those shows where all the right elements came together at the same time, meaning Tom and I were working together, Erik and Ethan got involved, and Steve Morrow, Jason Ruder, and Kira Roessler were involved. We’ve all worked together before, even before A Star is Born.

This was one of those once-in-a-career mixes where everyone was together all the time, and excited all the time.

The project lent itself to be super creative and everyone was on board with the same sensibilities and ideas. Todd gave us the latitude to be bold. This was one of those once-in-a-career mixes where everyone was together all the time, and excited all the time. The schedule allowed us to be bold and creative. It was really unique.

TO: It feels a bit like capturing that magic take of a band playing the song that they’ve played so many times but that one particular performance just has this magical quality to it – like, you’ve never heard it like that before. That is the take and for us, it feels like we got that on this movie. We’re just super happy with it.

A big thanks to – for giving us a behind-the-scenes look at the sound of – and to Jennifer Walden for the interview!

Please share this: