This episode originally aired on The Jordan Harbinger Show

What does a sound designer do all day? How do noise canceling headphones really work? And why do modern movie trailers all sound the same? In this follow-up to our 150th episode, Dallas shares the secrets of sound design with Jordan Harbinger. They also discuss audio deepfakes, sonic branding, and the eternal argument over analog vs. digital. This interview comes from The Jordan Harbinger Show.

MUSIC FEATURED IN THIS EPISODE

The Best is Yet to Come by Will Patterson
Find Yourself by Brightout
Never Land by UTAH
Car Chase by HAMPUS NAESELIUS

Follow the show on Twitter, Facebook, & Reddit.

Subscribe to our Youtube channel here.

Become a monthly contributor at 20k.org/donate.

If you know what this week's mystery sound is, tell us at mystery.20k.org.

Subscribe to The Jordan Harbinger Show wherever you get your podcasts.

Download Overcast in the App Store for a powerful, simple podcast player.

Visit athleticgreens.com/20k to get a one-year supply of vitamin D and 5 travel packs free with your first purchase.

View Transcript ▶︎

[music in]

You're listening to Twenty Thousand Hertz.

We just reached 150 episodes, and I still can’t believe we’ve made it this far. To celebrate the occasion, I wanted to pull back the curtain a bit.

Last week, you heard a deeply personal story about my life before this podcast.

This week, you’ll hear part of a discussion I had with my friend Jordan, over on his podcast the Jordan Harbinger Show.

Now, this episode is gonna sound completely different than our normal show. It’s a lot more conversational, and very fast paced. But Jordan’s a fantastic interviewer, and does a really great job of asking thoughtful questions that people ask me about often.

In this episode, we cover a lot of different topics… what sound design is, how it’s used in movie trailers, whether analog sounds better than digital, deepfakes, sonic branding, and more. It was a really fun interview and I hope you’ll enjoy it.

Here’s Jordan.

[music out]

Jordan: I'm not sure people have heard of sound designers before, and if they have, they're probably on the same page I was before this, which is, oh yeah, they pick the music for a scene in a movie [sfx: jazz music] and that's about it. [sfx: cheesy romantic music] Or maybe they add like a punching sound if the punch isn't loud enough in the stunt scene. [sfx: over the top punch]

Jordan: But obviously there's a lot more to it. And you're kind of part of this big magic trick the audience isn't thinking about what they hear, but what they see. And if it were a drawing, you'd be shading it right. With a pencil.

Dallas: You're spot on because I think a lot about sound design as a slight of ear, because we're essentially putting up a smoke screen for what it actually sounds like. And so what a sound designer actually does is just tons of layers of stuff.

Dallas: I do a lot of trailers with big bwahs [sfx: bwah] and boojs [sfx: booj] and things that just get you totally hyped [sfx: trailer drop].

Dallas: But there's just like a million sound effects, like categories and layers to every single piece of video or film or television or ad that we do.

Jordan: So when you look at two guys in a trailer they're like, all right, the trailer is, they're climbing a mountain and it's really isolated and it's really snowy.

Jordan: Are you thinking in your head, “We're going to make that whistling sound, that snow hitting snow on the top of a mountain makes.”

Dallas: Yes. Because wind doesn't sound good at all. Wind, if we just take like a sample of wind [sfx: wind] it sounds like white noise. So even you saying like “Top of a cold mountain,” I'm like, “That is like a fake wind.” So every time it's cold, I'm like, “Don't give me like the white noise wind [sfx: white noise wind transition to whistley wind] I need the whistley like performed wind that's like shhhhhh,” and kinda perceive it that way as humans. But when you put a microphone somewhere, sometimes the microphone does not capture what your brain has interpreted. It just sounds like woooosh, and there's really like no character or coldness to it.

Dallas: We can choose and craft these sounds and mold them to have a character of their own. And so even like with every environment that we do or foley or something, it's all performative, we're thinking a lot about what's being said, should it be dissonant wind? Is something dissonant happening on screen? So we'll even like affect choices based off of the content of what we're trying to make the viewer feel like.

Jordan: One thing I was thinking of was uh I think it was free solo or maybe it was another movie that I saw where somebody was climbing and there's a scene where it might've even been like Mission Impossible with those fake rock climbing scenes with Tom Cruise.

[sfx clip: Mission Impossible II]

Jordan: But there's one, you know, he's hanging with one hand and I noticed the wind got higher pitched, [sfx] which means faster. I'm like, “Oh, so the wind just happens to kick up even faster when he's hanging by one hand.” And I thought, “Oh no, no, they're trying to get us to feel like he's off balance and that things are happening rapidly.”

Jordan: So they just speed the wind up. In reality, there's probably no wind on this area at all cause his clothes aren't even really moving or anything. He's just hanging.

Dallas: Yeah. It's like a magician. What a magician is doing is something that looks so eloquent and clean in front of you, but there's all this dirty work that happens, just in the shadow. You can't see it and everything looks really fluid. Everything makes sense. But like right underneath the surface, there's a lot of work happening just to craft every microsecond.That's sound design, essentially.

Dallas: Sound design is very much like composing the real world in a lot of ways. You know, when we've also had this rich cinema history, that has taught us what these things sound like as well, which are not really real.

Dallas: Like when we are doing footsteps on grass, most of the time, you know that like the little like yellow sweeper, it looks like there's, it's almost like hay. It’s that it's just the real standard, like broom. Most of the time when we're just putting in footsteps that are on foliage or grass or whatever, we're just crunching that as we're watching the screen [sfx: foliage footsteps], just because it sounds so much more distinct. And a lot of times, actually with dirt, a lot of foley studios will just have a little dirt pit where I've seen him manipulate just little, tiny bits of dirt with their fingers, [sfx: dirt] because it sounds more crisp and clear and also dirt is dirty. So like you don't wanna just start stomping around in a room and kicking up dust.

Jordan: For people who don't know what foley is this, this is… Well, how do you even explain this? I've seen it in studios. I'll rent a studio and I'll go, “Why do you have gravel in a small kitty litter sandbox looking thing? And they'll go, “Oh, we do foley here too”. And it's just making a sound that is supposed to be like footsteps hitting the ground. Like you just mentioned with dirt, except you have to do it into a really high quality microphone. And like you said, you don't want to just start making a huge mess in a room with a $30,000 soundboard. You have to have it sort of localized.

Dallas: So foley is essentially the sound elements that a human in the piece is touching. Whether it's with their hands, it could be their clothing, and it could be their feet generally. Now foley will kind of veer into more unique sounds and stuff because they have giant rooms of junk. But for the most part, the core aspect of foley is done by humans in front of a screen in real time. Because what they're trying to do is all of those intangible moments they're trying to perform for the actor. It's like an extension of what the actor is doing. So if you have bad foley, you can actually take away from the performance of the actor.

Dallas: I think that the confusion with foley is like, “Why don't you just go to a library, find footsteps and then kind of sync them in there?” Foley's job is to be a very performative art because if you just put in stock footsteps [sfx] even the way we walk can vary greatly. Like if you're mad at your partner or something and you want to stomp away, [sfx: stomp away, door slam] that's a different performance of a stomp than I don't know someone, just uh, you haven't seen in a long time arrives at your front door [sfx: Footsteps, open door].

Dallas: So, you know, we've talked a little bit about like environments, but there are these kind of funny sounds that they're kind of funny to me because they're so overtly trying to like affect you.

Dallas: I’m thinking of things like in a trailer where you hear like a bwah or it's like bwah , you know, “Are we ever going to get there?” bwah [sfx] “Maybe” booj [sfx] and then some like a line that's like a little piece of what's going to happen in the movie followed by like a whole re-imagined eighties track in a new way. That's kind of spooky. It's like the building blocks of a trailer. And so like hits [sfx:hit] and shimmers [sfx: shimmer] and shings [sfx: shing], and those things are used just so liberally in advertising just to like reach into your like primal brain and try to pull on that fear response.

Jordan: As you described that, I'm like...

[music in]

Jordan: Okay, characters, standing on a open plane, dystopian skyline in the distance flashed to like split second of an action scene where an alien is like reaching out towards [sfx] them later on in the movie. Right Exactly. And there's drum beats and that loud. Yeah. That loud, like foghorn, that's not quite a foghorn thing. That's just sounds like. Yeah. Yeah. Yeah. [sfx: bwah]

[music out]

Jordan: And so it is funny cause it's almost predictably cliché. And yet if you find a trailer that doesn't have that y'all, didn't do your job, right, in this decade anyway.

Dallas: Yeah. exactly.

Jordan: Until we get sick of that, because everyone uses it and then you have to find something else.

Dallas: But there's a good reason for it. And this is what's so fascinating just about sound design to me, is like a lot of this stuff may seem like “Oh, that's a scary sound let's put it in.” But a lot of this stuff is very much psychological. It's playing off of our primal urges and feelings.

Dallas: We're living in like the future right now. This is our ancestors would see this, like we're living legit in the future.

Dallas: Now, even a couple generations ago, the lifestyle for all of human history before that was a lot different, you know, hunter gatherer, basic needs, things like that. So like when we put in these hits, like our brains are still living in a thousand years ago you know, we're, we're getting there, but there's so many things in the way that we do things that just our brains aren't prepared for. Anxiety situations, you know, public speaking, things like that and sound, that's a thing that you can kind of play off of.

Dallas: You know, if you're in the creaky old house and you hear one little twig break [sfx: twig break, house ambs, bulding horror sfx] or something in the middle of the night, like your brain goes straight to, “There is a robber and I'm in trouble.” Meanwhile, it was just like, your house settling or something like that. Like a lot of those sounds are really just trying to hit you in those places that are kind of the primitive brain to make you feel these like early responses, that they would naturally have.

Dallas: And so it is very much like a magic trick mixed with neuroscience. Now I argue that we've gone so far in the epic direction, especially with like superhero movies and things like that, that it's becoming a super cliche thing.

Dallas: even if we rewind a decade or two ago, do you remember when trailers were just like [sfx: trailer music] Jack met Julie in the coffee shop and there's a little bite.

Jordan: Yeah. It's just stuff that happens in the movie. That's cut together.

Dallas: It's just stuff that happens cut together with a narrator. Now we've hit this, like the 2010 plus cliche. I think people are getting a little tired of, which is just like the overly epic everything. Rather than crafting a mini story or whatnot. But you know what, it's marketing, there's a lot of money involved, we're trying to get $500 million out of a movie. they're going to probably throw everything they can at it.

[sfx: trailer hit montage]

[music in]

Coming up, I give my take on some of the biggest sonic questions out there. Questions like, is analogue really better than digital? What makes a great sonic logo? How do noise canceling headphones actually work? And how will society deal with the rise of audio deepfakes?

That's all coming up, after the break.

[music out]

MIDROLL

[music in]

We just reached our hundred and fiftieth episode. To mark the occasion, I sat down with my friend Jordan Harbinger to talk about sound design, movie trailers, this podcast, and more. Jordan also asked me about some of the most important questions in sound today, including the never ending debate over analogue versus digital.

[music out]

Jordan: When people say analog sounds better than digital, is that kind of the case?

Dallas: I am not really in the camp that analog sounds better than digital, but what I will concede is that the way that we approached music in the analog era was much more active than how we approach music in the digital era. And what I mean by that is like, whether you had a cassette tape [sfx: tape deck clicking shut] or a record [sfx: needle drop] you had to dig through bins to find something that you may or may not even be able to listen to.

Dallas: You look at the artwork, You go through it, you pull it out, you go buy it. You're invested in listening to this. You have like skin in the game for this thing. So then when you put this record on you're like training your body and mind to think about it and engage with it. You only had so many of them, I could only have, a few of them and some of them I didn't even like, but I would listen to it cause that's all I had.

Dallas: The thing that we lost in the transition from analog to digital is that approach because now it's like, everything is so disposable. If I don't like something in two seconds, I can skip [sfx: skipping through tracks quickly].

Dallas: And so we get like a big series of big singles and music that's now being made for that style of just like almost click baity type. I mean, there's still amazing music being made, but I think that's the thing that like bums me out the most is that we just don't approach even listening to music with the same level of skin in the game as we did when they were analog mediums.

Dallas: So I wouldn't really say they sound better or worse. I just think that our minds were more conscious about listening. Therefore, that's a big reason why I think analog sounds better because we're more primed for it.

Jordan: Looking at new devices, like noise canceling headphones, which I really like, they create a bunch of different sounds probably that I can't hear that fight... well, actually, how do those work?

Dallas: The simple answer is that there's a little, little microphone. That's recording everything outside and it's coming in and it's instantaneously flipping the signal.

Dallas: So think about like what you would consider like a sine wave, or just for anybody who's like, have you ever seen just like audio waves, like think of audio waves in your brain, you think of a line that goes up and there's a line in the middle and then that line crosses and it goes down and it comes back and meets in the middle.

Dallas: That's a sine wave. It's the most pure tone. [sfx: sine wave] So we're going to think about the sine wave. So what happens is, is if you're recording that sine wave, what happens in the noise canceling is immediately it's flipping the phase. So that's what it's generating, an opposite signal to the signal that's coming in.

Dallas: So therefore in theory, if you have a waveform going up and the same waveform going down, it cancels out to nothing. So that's, even if you've ever worked in audio or anything like that, if you just simply take a piece of audio, put another piece of audio, flip the phase on it, it will go to silence but yeah, that's essentially how that works.

Jordan: So if I'm canceling out a really loud sound, does that mean I'm just playing another really loud, but opposite sound into my ear.

Dallas: Yeah. There's a lot of debate on that. I've talked to Bose directly about that specific thing. The idea is that you still don't want to be slamming doors around noise canceling headphones, because what you'll hear is the noise canceling will fail. So a lot of times, like if I have headphones on and maybe I slam a door [sfx: door slam], you'll actually hear it disengage, and you'll hear like, uh, just kind of a clicky digital thing [sfx: clicky digital thing], because it just can't keep up with that level of volume because it's so fast.

Dallas: It's actually doing a cancel out by the time it's hitting your ear, but at the same time Bose, and these companies would say you don't use these for hearing protection. So I'm sure there's a lot of nuance to that.

Jordan: What do you think of deepfake sound? This is fresh on my mind because I just saw that Zelensky, the president of Ukraine. There was a Russian deepfake of him saying, “Oh, we're surrendering, lay down your weapons”

Jordan: And you know, “The Russians are gonna take you prisoner and it'll all be good and fine.”

[sfx clip: Zelensky deepfake]

And it didn't fool anyone, cause the Ukrainians have been seeing Russian disinformation for a really long time. And so has it all of Eastern Europe for that matter. But they can do pretty impressive, deepfake voices. The video's not going to be too much further away in terms of processing power and the ability to do that in the next few years in a convincing way.

Jordan: But these deepfake voices…I haven't seen what Adobe has said they have and shown what they have because they won't release it to the public. Right? For good reason, because it's just going to create all kinds of chaos.

Jordan: I predicted that they were going to make a politician either say something that they would never say that's either racist or a big gaff or a policy change. I didn't think they would have a president surrender in a war, but it's the most obvious use for this kind of technology. But with vocals, it's really going to be hard to detect.

Jordan: Can we trust our ears now?

Dallas: At this moment, maybe. So I did a show on this and I tasked the team and I was like, I want my voice for the first minute, minute and a half to be as convincing as possible, but it's not me.

[sfx: film beep]

[sfx clip: deepfake Dallas]

Deepfake episode dialogue: Wait a minute, what’s going on here? Deepfake episode dialogue: I was in the middle of saying something. Deepfake episode dialogue: Sorry, but who are you? Deepfake episode dialogue: I’m Dallas Taylor. Deepfake episode dialogue: Uhh, no, I’m Dallas Taylor. Deepfake episode dialogue: No I think you’ll find, I am Dallas.

[sfx: film beep]

Dallas: And so to do that, we worked with one of the guests. and the guest even was very concerned about telling us exactly how it was done. Like we never really got an answer of the actual ways it was done, but essentially he had said, “Send me three hours of your narration, just nothing else. And then I'm going to push it through some supercomputer and it's going to do a lot of AI stuff and machine learning. And then that's going to make this thing and the more audio, the better, because there's just all this nuance.”

Dallas: And it just takes a long time to make it convincing. So we did that, fooled a lot of people with the intro not being me. And so the experiment was like, “Can I convince people who've been listening to all of these shows that I am not real?” And I think it did a good job. The show is called deepfake Dallas, so like you kind of are tipped off right off the bat. So I think that you can kind of get it, but if you're just, if you didn't even look at the title, I think it's really convincing . Even when we make episodes, now we put in a deepfake voice of me. We have the words per minute, my cadence, all that stuff that we can build out this show, then I can approach it as an outsider and then craft it before I actually come to it.

Dallas: So there's a really interesting uses for deepfake like that. I would never use it on the show, but it's just like a tool to put me in the seat of the listener rather than me being the seat of the host, which is a different perspective.

Dallas: But will it get better? Yeah, it's gonna, it's gonna keep getting better. I haven't heard a lot of deepfakes that are super convincing yet. Where the pitfalls are is in emotion. And I know that's the next frontier. Where when I'm speaking right now, I'm going up, I'm going down, everything has meaning, I'm communicating, meaning in the way that I'm singing.

Dallas: So that's like the difference between what the machine is doing now versus where it's trying to go. And can you get the machine to understand the context and emotion? I think that's going to be the thing that makes that more convincing, and it will be able to pick out in the source audio when someone's happy or the quirks that they do.

Dallas: I mean, I have voice quirks, you have voice quirks that you just know, are you. Yeah. So like the way that we sing our words is complex and will it get better? Yeah. It will.

Jordan: Yeah. To the point where it can be used against us, especially when you're talking about nation states. Like, I don't think a nation state like Russia or Iran or the United States would have any qualms about grabbing Adobe software and being like, “This is what we're going to do with this. We're going to make the Ayatollah say this horrible thing, and we're going to play it for everybody in Iran. And it's going to change the political system over there in a way that we can have that's desirable for us.” That is almost for sure already happening, but it's undetectable, or we're very close to that point. And we won't know until later that something is fake.

Dallas: You know, one thing I find interesting is that social media and how much that is important to politicians now, I think is going to be a combating effect of that.

Dallas: If you have the talking head in front of a podium and that's all of the communication comes from, the White House pressroom. [sfx: White House Press Room welcome] You have a lot of data that you can make sound just like that. But when you have someone like Zelinsky or whatever, going out and just, showing him doing this and doing that and being in the world and just a, terrible microphone on the phone or whatever, that's going to be protection with deepfake, because that's going to be hard when you're trying to not only get the cadence right. And all this stuff and edit.

Dallas: That's the other thing that we forget about deepfakes, I'm a sound designer. I'm a sound editor. I can not only get a deepfake, but I can adjust it in ways and then edit it to make it exactly the way that I want. I mean, that's what we did in our show. We did a deepfake, but we did like five takes of deepfakes and still pick the best ones.

Dallas: So I think that like part of the way to combat that is putting yourself in situations where it's just not as easy to do that. [sfx: outdoor ambs] If you're out in the open, if you're social media style, like that's going to be detectable. There's a lot of complexities in sound design, to not only deepfake the performance, but then also get the environments right. Because I think a lot of experts, even like me, I think I'd be able to pick out most if I really analyze how it sounds.

Jordan: Sonic branding is interesting, right? The second I give you an example of the Intel inside sound, right?

[sfx: Intel logo]

Jordan: When you see those commercials. Everybody knows what that sound is. The second they hear it, they're looking at the logo in their head.

Jordan: And so this is a brilliant play because you can evoke a branded logo or animation in someone's head with a sound, even if there's no screen anywhere near you. What else is going on with Sonic branding? That can't be the only reason they do this.

Dallas: When Netflix, leaned into that and made a sonic brand that was there all the time, I felt like that was a moment in history where it really started to explode. Many more companies started going, “What's the reasoning for this?”

Dallas: You know, we're very visual creatures in general. So like a logo seems to make a lot of sense. Artwork makes a lot of sense.

Dallas: With sound, we just don't think about it a lot. So we don't go, What, what does it matter what something sounds like?”

Dallas: If you walk into like your grandmother's house and you smell that dish that she only she makes [sfx: sniff… ahh] or that baking that only she did. That just washes you with nostalgia and good feelings. So when you think of something like Netflix, which I consider the most iconic sonic branding out there, it's a call to action that says “You're done. Put your phone down. You're about to watch something incredible.”

Dallas: Your brain unconsciously is thinking about this because you know, hopefully you're watching stuff that has a consistency of entertaining you. And if you have that positive association, then that sound is essentially training you to go. It's time to stop with all those other things.

[sfx: Netflix ta-dum]

Dallas: What was interesting about that particular thing is when I was, it was actually the first time that the company had ever spoken on it. So I was speaking with Todd Yellin, who had led the effort at Netflix.

Dallas: They were exploring all these different options, and when they were kind of putting this in front of audiences. There was another option that I found hilarious is that they wanted to kind of evoke the old Leo the Lion idea, you know MGM.

[sfx clip: Leo the Lion]

Dallas: They wanted this to feel like a big cinematic moment. So they thought a lot about animals being a part of it or having some sort of call and response in there. The thing that I learned out of that logo that just kind of blew my mind is he said, “There was almost a goat bleat in that sound.” And I was like, I don't believe you. And he was like, you can hear it. .

Dallas: The PR person played it straight off of their phone. They said, no joke. This is exactly what we were doing. And it sounded exactly like the Netflix logo. It was just like this.

[sfx clip: Netflix goat]

Jordan: The ingredients of this are really interesting. I don't understand for the life of me, how people come up with the idea on how to make a sound, but that's why you guys do what you do, because it's not something that I can just bet anybody can just pick up and do randomly.

Dallas: I think there is this assumption that a sonic brand is a cool sound. That is just the opposite. I would say that the thing that takes us the longest out of everything we do is sonic branding. The shorter the sound, the longer it takes, and this thing can take months. And the reason it takes months is because every one of these sonic brands means something deeply.

Dallas:. That's one of the reasons I started the podcast is because there are these incredibly nuanced stories behind why things sound the way that they do that are just going to be lost to history.

Jordan: Dallas Taylor, thank you so much, man.  Dallas: Thank you.

CREDITS

[music in]

This episode was part of a longer conversation I had with Jordan. We also talked about jingles, ASMR, movie dialogue, and more. You can hear the full interview on The Jordan Harbinger show. Over there, Jordan talks to all kinds of fascinating people, from legendary artists to intelligence operatives, to star athletes. Recent guests include basketball phenomenon Dwayne Wade, psychologist Steven Pinker, and journalist Anderson Cooper.

In these interviews, they break down exactly what they do, and how they do it. Jordan is a great interviewer, and it’s always fascinating and insightful. So go take a minute to go subscribe to the Jordan Harbinger show, right here in your podcast player.

Twenty Thousand Hertz is produced out of the studios of Defacto Sound. For some tasty sonic snacks, follow Defacto Sound on Instagram.

Jordan: This show is created in association with PodcastOne. My team is Jen Harbinger, Jase Sanderson, Robert Fogarty, Millie Ocampo, Ian Baird, Josh Ballard, and Gabriel Mizrahi.

Before we go, I want to extend a huge thank you to all of our listeners for supporting us, engaging with us, and helping us get this far. Your support and enthusiasm really means the world to us, and we couldn’t have done it without you. Here's to another 150 episodes.

Thanks for listening.

[music out]