Deep Dive Into the Thrilling Sound of ‘Star Wars Jedi: Survivor’

Star Wars Jedi: Survivor – developed by Respawn Entertainment and published by Electronic Arts – is the highly anticipated sequel to Star Wars Jedi: Fallen Order (2019), and picks up five years later with young Jedi Knight Cal Kestis continuing the fight against the Galactic Empire. Here, the sound team (led by Audio Director Nick von Kaenel at Respawn Entertainment) talks about how they built upon the foundation of Fallen Order while still creating something fun and fresh for Star Wars games fans. They talk about designing bolder sound for the different colored lightsabers, creating franchise-fitting new sounds for foes, world-building for unique environments, building systems for breaths, foliage, ships, combat, debris, and Force abilities, using blueprints in UE4 to add audio to game logic, working with Wwise middleware for spatial audio, and so much more!

---

Interview by Jennifer Walden, photos courtesy of Electronic Arts

---

Please share:

• More

Respawn Entertainment’s Star Wars Jedi: Survivor is the action-packed sequel to Star Wars Jedi: Fallen Order (2019), continuing the story of Jedi Knight Cal Kestis as he travels to new places and takes on new foes.

Jedi: Survivor offers gamers a fresh take on the Star Wars franchise sound while still honoring what’s come before in the IP. For instance, Respawn’s Senior Sound Designer Oscar Coen talks about taking a bolder approach to sound for the different colored lightsabers, and Audio Director Nick von Kaenel talks about researching Sound Designer Ben Burtt‘s original approach to the Rancor for Star Wars Episode VI — Return of the Jedi to find inspiration for creating new sounds for it in Jedi: Survivor.

Here, von Kaenel and Coen are joined by other members of the Respawn sound team – Senior Sound Designer Alex Barnhart, Sound Designer Ashton Faydenko, Sound Designer Kartika Luky, and Senior Dialogue Editor Nicholas Friedemann – and Senior Cinematic Sound Designer Tom Jaine (now a Supervising Sound Editor at Sweet Justice) to talk about their sound work (including recording source material) for the game’s AI combatants and companion driods, creatures, locations, and weapons. They discuss their use of Wwise middleware for spatial audio, Wwise motion for advanced PS5 haptics, using Wwise for callbacks, loops, and effects processing, working with UE4’s blueprints, creating new systems for Force abilities, breaths, foliage, ships, combat, and debris, and so much more!

Star Wars Jedi: Survivor – Official Story Trailer

Star Wars Jedi: Survivor is built upon the groundwork of Star Wars Jedi: Fallen Order. In terms of sounds, what did you learn from the first game that helped you to improve (or expand) the sound of Jedi: Survivor? What were some aspects of the sound that you wanted to change? What did you want to keep?

Nick von Kaenel (NVK): I was satisfied with the direction we took the sound on Star Wars Jedi: Fallen Order. We were careful to keep the sound design in the same direction as the films, but we also found ways to make the game sound unique and fresh while supporting the story and gameplay. I think we struck a good balance, so I didn’t want to stray too far from that.

We were careful to keep the sound design in the same direction as the films, but we also found ways to make the game sound unique and fresh…

The goal with Star Wars Jedi: Survivor was to continue with this established approach but refine it further. During the early stages of the project, we looked at all of Cal’s sounds — from the lightsaber to his foley and his Force abilities — and looked for any improvements that could be made. We updated assets to make them punchier or sit better in the mix. We added new systems to make the Force powers interact more with the environment or context. We revamped the foley and added more detail to all of the movements. We also added more real-time processing and mix behaviors to make sure that the important sounds were always cutting through when needed.

We also refined all of our audio tech. We implemented Wwise Spatial Audio, setting up rooms and portals for all the levels in the game. We made improvements to the audio propagation pipeline so that it felt right with the third-person camera and also avoided bugs with the camera/player clipping through audio volumes. I refactored all of our attenuation sharesets so that they felt more realistic at various distances. I also made carefully-tuned shared reverb buses that could be used in all of the levels, so that we had consistent reverb levels and quality. We implemented Wwise motion in order to make use of the advanced PS5 haptics, and implemented 3D audio for all platforms. We also built tons of smaller systems to help with implementation and mixing, like ways to position or play the audio depending on the game objects, or ways to control RTPCs and mix states based on the player’s actions.

I also made carefully-tuned shared reverb buses that could be used in all of the levels, so that we had consistent reverb levels and quality.

Another goal with Survivor was to have more intentional focus on the player experience. With games of this size, it can often be difficult just to get coverage on all of the content. This can result in sound being added without much thought about the greater picture. We wanted to make sure that we were being thoughtful about our choices, finding opportunities to help tell the story and intentionally define the player experience. We wanted to consider how the player is going to experience our sounds, and what kind of emotional response we are conveying.

Oscar Coen (OC): In general the biggest challenge with Jedi: Fallen Order was just figuring out how to keep everything feeling authentically Star Wars while also sounding modern and high fidelity in a way that would work well in a game context. After seeing positive fan reception to some of the out-of-the-box things we did sound-wise in the first game, we realized we had a lot more flexibility with how far we could push things while remaining authentic to the franchise than we initially realized.

…we realized we had a lot more flexibility with how far we could push things while remaining authentic to the franchise…

This time around we tried getting a lot bolder with our choices and tried to focus more on being inspired by the original films rather than simply imitating them. This meant trying to really learn from Ben Burtt’s thought process and methods for sound design, particularly his use of musicality and tonality in his sounds, and applying that to our own personal style.

As another example of pushing things further, one big change from the first game is how much more dramatically different the lightsaber colors sound from one another (though in the fiction of Star Wars, the color of a lightsaber wouldn’t necessarily affect how it sounds). But similar to other sounds, I was really cautious about going too far with the processing on each color because I was already kind of going outside the norm with the core lightsaber sound and thought if I went too far with the colors it would just not sound like a “lightsaber” anymore.

…the rarer colors like Pink, Yellow, or White were made to sound much more different from a standard lightsaber sound.

Also, at the time, not as much new Star Wars media had come out that had really expanded on what different lightsabers could sound like, except for Kylo Ren’s saber. After Jedi: Fallen Order was released, I realized I had just become way too sensitive to the smallest changes in a lightsaber sound that I had made those color differences way too subtle — only the most attentive players would pick up on it. For Jedi: Survivor, I kept a lot of the same approach for each color but just cranked up the effects way more, still entirely dependent on the built-in Wwise effects for the swings and idles (the lightsaber activation sounds were unique assets). I also took more time to really dial in the mix for each color so the clarity is maintained no matter what color you use. The more common colors like Blue or Green were designed to only sound slightly different, while the rarer colors like Pink, Yellow, or White were made to sound much more different from a standard lightsaber sound.

Another big learning from the first game was mixing. In the first game, there was a lot of just trying to get things to work, so our mix was a lot more static, relying primarily on HDR for any kind of dynamic change. On Jedi: Survivor, I really wanted the combat to have clarity in its mix and to be very aggressively focused on gameplay sounds in particular, so much so that in theory people could play the game blindfolded and still know exactly what’s going on. This ended up leading to a wild web of RTPCs constantly turning things up or down based on different contexts, like what abilities the hero is using, or the number of enemies in combat — all with the goal of making sure you’re always hearing what you need to hear when it happens but also maintaining the satisfying sound of the core gameplay as well. The way we tested this was by very frequently playing the game and trying to also play it with my eyes closed to see if I could keep track of everything that was happening and still play well.

Star Wars Jedi: Survivor Audio Team and Composers

Can you talk about your approach to sound for the enemies/beasts/combatants in Jedi: Survivor? What were some unique sounds (foley, vocals/vocal processing, and other effects) you created for the:

Rancor:

NVK: I was thrilled to work on the Rancor for this game. I started by listening to the wonderful designs Ben Burtt had made and read a bit about how he created them. For the growls, he had recorded a neighbor’s small dog and pitched it down, so I figured I would start there. I recorded my dog Zoey, an Italian Greyhound, who loves to growl when playing tug of war and I did essentially the same process to get more content. The original Rancor has a ton of flanger on it, but I toned that down a bit to make it feel a bit more realistic.

I recorded my dog Zoey, an Italian Greyhound, who loves to growl when playing tug of war…

I tried to reverse-engineer the animals used for the roars and created new assets for all the various animations we had. With all the assets, I added a ton of bass enhancement and saturation to make it feel huge and powerful. All in all, it was a fairly simple approach, but we were happy with the results.

Sutaban Alpha:

Ashton Faydenko (AF): Early in my work with the Sutaban, I struggled quite a bit with finding an interesting palette for their vocals. As I hunted for inspiration, I eventually stumbled onto some early concept drawings of the Sutaban, finding some interesting tidbits in the process. They have these bulbous growths running from their heads to their back that were actually called out as ‘water sacs’ in an early paintover.

…this isn’t just an apex predator, but an animal that’s trying to hoard as much moisture as possible.

This ended up being the seed I was looking for, with every vocal element I designed working to reinforce that this isn’t just an apex predator, but an animal that’s trying to hoard as much moisture as possible.

I used recordings of slime, camels, and my bathtub draining to create gurgling liquid textures, while the whine of an infant horse provided some very alien-sounding top end. With how crisp and dry Jedha sounds, I really liked the strong juxtaposition this wetness created, helping the Sutaban to make a stronger impression in the final mix.

The Massiff (Magnaguard):

Tom Jaine (TJ): The Magnaguard was the first AI for me to work on. It presented a complex challenge, as the aim (as with much of the sound in Star Wars) was to seamlessly integrate the established audio identity from the films and TV shows with our game’s distinct sound design. Key inspirations included the tonal impact sounds created during Ahsoka’s fight with the Magistrate from The Mandalorian.

All original sounds were carefully cleaned and enhanced using Waves Z-Noise and iZotope RX Deconstruct to ensure pristine quality…

To create a rich, engaging audio experience, I meticulously blended the tonal sounds, the electricity source I created with Serum by Xfer Records, and the iconic electricity sound of the Magnaguard’s staff.

All original sounds were carefully cleaned and enhanced using Waves Z-Noise and iZotope RX Deconstruct to ensure pristine quality; many of the assets are over 20 years old, so they needed a little dusting off. One of the main standout aspects for the Magnaguard’s sound was the spinning attacks and ensuring their attacks were obvious to the player.

Bedlam Raider Droids (B1, B2, BX):

Alex Barnhart (AB): My first major tasks on Jedi: Survivor were designing the sounds of the B1, B2, and shortly after, the BX droids. It was important to make sure they felt unique while still feeling like they were in the same vein aesthetically.

…the main goal was to make sure that the B1s felt small and goofy, the B2s chunky and powerful, and the BXs sleek and deadly.

I started by finding any sounds that were available and already designed for them, both from the movies but also certain sounds (not including voices) from EA DICE on the Star Wars: Battlefront games, to maintain as much of the original DNA as possible. After getting the first pass in, the main goal was to make sure that the B1s felt small and goofy, the B2s chunky and powerful, and the BXs sleek and deadly.

As you find so many of these droids in the game, I made sure they had clear and distinct sounds for combat tells, like when they ready their blasters or wind up for a strike. This was critically important for our game as the mix can get quite chaotic and these sounds need to stand out so the player can know when to strike and when to block.

Mire Terror (Mogu):

AB: The Mogu was one of the earliest creatures I was tasked to work on and probably had the most revisions for me personally of anything in the game.

…I used a lot of marine recordings like seals and walruses for the main tonal layers of its vocals…

Being one of the larger creatures in the game you fight, the intent was to make the Mogu sound menacing and powerful. The Mogu takes a lot of inspiration from the Wampa both in appearance and its sounds. After reading up on what Ben Burtt used to make the Wampa, I used a lot of marine recordings like seals and walruses for the main tonal layers of its vocals and combined that with recordings of my own voice, slime, and other animal vocals.

E3-VE3 (Droideka):

AF: As a huge fan of the prequels growing up, taking on the Droideka was equal parts intimidating and exciting! They don’t actually show up too often in the films or TV shows, which meant there wasn’t much of a pre-existing palette to work from, especially when it came to their iconic bubble shields.

…there wasn’t much of a pre-existing palette to work from, especially when it came to their iconic bubble shields.

While a lot of the ‘Star Wars sound’ is rooted in classic ‘70s tape effects, the prequels definitely started to explore digital processing a bit more, which was something I wanted to play up here. Almost every layer of the bubble shield is a unique Kilohearts Phase Plant or Serum patch, outside of a few heavily distorted packing tape and plastic wrap recordings to give the electrical elements a little more of an organic feel.

D-L1t (DT Sentry):

Nicholas Friedemann (NF): During my research phase into how we were going to process droids in Jedi: Survivor, I compiled as many examples of each droid talking as I could. The DT Sentry had a thimbleful of examples of speech in Star Wars Rebels but the lines weren’t meant to be discernable. For our game, the lines needed to be clear so copying what was pre-existing wouldn’t work for us. I had to make something new.

…I found that most of the droid processes amplified a lot of mouth noise, clicks, and sibilance.

Matt Wood provided the voice and gave us a read that was similar to the B2s which was perfect. I edited all the VO to be as clean as possible because I found that most of the droid processes amplified a lot of mouth noise, clicks, and sibilance.

Once the files were squeaky clean, I applied the process. I wanted something reminiscent of the B2s but bigger. Much bigger. There is some pitch shifting but the bulk of the effect is iZotope VocalSynth 2’s Talkbox, which gives the lines that same burpy, pitchy quality of the B2s.

Overall, what were your biggest challenges when it came to designing foes?

NVK: AI is always a huge challenge. There are a lot of things to consider aside from just the creative design of the sounds. In addition to the creative aspects of designing the sound, you also need to keep in mind how the AI fits into the greater mix. Are the gameplay cues readable? Does the AI sound match its power level? Do the sounds work if you are fighting multiples of the same AI? There is a lot of iteration to get things feeling right.

What were some important aspects of the sounds for the different locations in the game? How were you able to use sound to make these locations feel unique:

Coruscant:

NVK: There are a ton of great sequences in Coruscant, but aside from that, the ambience of the level was heavily defined by the sounds of the ships passing through and also the various billboards and advertisements.

We worked with Overworld Audio, who created a ton of characterful jingles and loops for the advertisements. We wanted to make it feel alive as much as possible, but also have a dark, downtrodden feel to it.

The sheer number of ships flying around meant we had to create some systems to keep the emitter count in check…

The sheer number of ships flying around meant we had to create some systems to keep the emitter count in check, but also have them feel alive as they pass by. I created over 100 different ship passbys which would get triggered so that they would doppler by at the exact time they passed the player listener.

For distant ships, the passbys would get folded down into a bed with random one-shots. We also had different assets that would play if you were inside and hearing ships through a window or door.

Koboh:

AB: Koboh is the largest planet in Jedi: Survivor and has a wide range of locations and biomes that you can discover there. It was important to make sure that even though you are on the same planet, when crossing the threshold from one level to the next, there was a very clear sonic shift. Luckily, the art in our game is incredible and all the different areas of Koboh have very distinct looks to them that we could draw on for inspiration.

I used a lot of the traditional methods…but found granular plugins to be a huge help in getting really odd-sounding creatures.

Early in development, we had some rough map ideas and concept art for what Koboh would look like and I spent a lot of time cranking out tons of variations on different ambient creatures, insects, birds, and more that might live in these regions and putting the recordings in our library for later use. I used a lot of the traditional methods, like pitching and reverse effects, but found granular plugins to be a huge help in getting really odd-sounding creatures. A lot of Star Wars ambiances always have unnatural elements to them and we often try to add some elements that just sound a bit weird, and these processes helped get a lot of that 80% of the way there.

The first time you land on Koboh, you crashland in a mostly abandoned canyon environment called Dredger Gorge and my intent was to make it feel incredibly desolate and almost hostile. The fauna you find here are rugged and the ambient birds and mammals you hear in the distance are mostly processed scavenger birds and coyotes.

While the valley is still a bit desert-like visually, we intentionally added many more songbirds and other more pleasant sounds.

To contrast this, when you find yourself looking over the valley into Rambler’s Reach, the ambiance changes quite a bit. While the valley is still a bit desert-like visually, we intentionally added many more songbirds and other more pleasant sounds. The intent was to make Rambler’s Reach almost pastoral by comparison. I can’t take all the credit for this; a lot of this work was originally spearheaded by Chris Walasek, and Kartika Luky took over some large parts of Rambler’s Reach as well.

The swamp and the subsequent Lucrehulk were other areas I was in charge of that had a drastic shift in tone and art style. The biggest thing I wanted to accomplish was having a general feeling of uneasiness and lean into as much of a horror game feeling as I could while still feeling like Star Wars. This meant that all the fauna in the backgrounds had to be eerie and unsettling and the Lucrehulk itself needed to sound like it would fall apart at any moment. The Shattered Moon had a lot of areas of big metal and it was important to make the Lucrehulk feel hollow and empty compared to the denser factory of the moon. Mixing in close-sounding creaks with huge, resonating rumbles through the hull of the ship really helped sell the size of the space.

Jedha:

AF: Jedha is one of the larger planets that the player can explore. However, since it’s almost entirely a desert, we knew that a lot of the player’s time was going to be spent listening to sand and wind. One of the biggest challenges we ran into was finding ways to make sure the various desert ambient beds had their own sonic character.

…I tried to achieve that…by making sure every ambient bed had both a distinct fundamental frequency and a unique textural element.

One of the ways I tried to achieve that was by making sure every ambient bed had both a distinct fundamental frequency and a unique textural element. Some beds might be tuned lower with strong buffeting storm winds on top, while others might be higher pitched with more sand whipping around. By making sure these beds all occupied distinct ranges in the frequency spectrum, we could get more variation from less content, either by combining them or transitioning between them in different ways.

OC: As Ashton mentioned, the biggest challenge with Jedha is that it’s a massive desert, and deserts tend to not sound very interesting naturally, so a lot of what we had to work with was wind and sand.

Ashton did a lot of the heavy lifting by filling in the natural details of the world and I spend most of my time trying to add more stylized or unique additional sounds to add a bit more interest to the soundscape. The main way I attempted this was with fauna, of course, but also by focusing on how different objects could theoretically react or resonate against wind and have tonal content emanating from them.

…giant metal decorative panels…have an almost musical metallic resonance that is dynamically controlled by an RTPC that tracks the overall volume of the wind beds in the level.

A good early example is when you first arrive at Jedha, there are these giant metal decorative panels throughout the level that have an almost musical metallic resonance that is dynamically controlled by an RTPC that tracks the overall volume of the wind beds in the level. This let me create the illusion of the metal panels resonating from the wind blowing through them by keeping them closely related to the movement in the beds. Of course, these were also designed to play well with the music and naturally blend with it.

This same approach was taken even further in a later part of the world where you reach an area called the “singing ruins.” The Narrative team really wanted to lean on sound for a lot of the narrative beats in this level, which was a really fun opportunity for the music implementer for the level Colin Grant, the composers, and I to work together on a level where sound got to take center stage.

…those pillars were designed to vaguely resemble a wind instrument.

Throughout the level, there are these large pillars you pull out for different platforming segments, and those pillars were designed to vaguely resemble a wind instrument. They played these loud, constant, wind instrument-like tones that would always be in tune with the music playing by making sure all the compositions in the area maintained a few common tones which I would design my sounds around.

This all leads up to when you acquire the Dash ability in this large cave where you pull out several of these wind pillars that all build up tones into a chord, which then resolves with the music that begins to play when you complete the puzzle, creating this cool moment where sound and music begin to blend together creating an almost dream-like state in the game where the line between diegetic and non-diegetic music gets obscured.

As for the fauna, most of it was me searching up the sounds of different desert animals, poorly imitating them with my own voice, then pitching my voice up or down enough so it stops sounding human.

Kartika Luky (KL): I worked on Cere’s Base on Jedha. One of the things that I learned from the team when working on levels is how the ambience sounds could help tell the story. In this case, Cere’s Base is a hideout and safe place, so the approach for the ambience is more on the quieter side. Most of the emitters I put there sound very distant, subtle, and quiet. I intentionally didn’t put too much rock stress or debris to avoid the area sounding like it was falling apart.

I intentionally didn’t put too much rock stress or debris to avoid the area sounding like it was falling apart.

As the player gets further in the game, they reach the part when they play as Cere and fight the Empire. This moment is actually using the same level and same space, but the situation now is completely different because the Empire is trying to breach the base, so we had to change the ambience too. There are lots of attacks and damages, so we put in lots of burning metal stress and creaks, some fire loops, crumbling rocks, and debris. The beds also contain some distant blasters and battle sounds, so the ambience sounds way more chaotic.

Alex Barnhart did a huge chunk of work on the siege as well, like the explosions and alarm sound to add more tension and a sense of danger, and did a lot of other creative and technical work.

Shattered Moon:

KL: Shattered Moon is divided into two visits. The first visit goes up until when you acquire the Crossguard stance, and the second visit involves entering the laser tube, going to the lab, and fighting Rayvis. Both visits have completely different feelings and the levels look different visually.

When the player first enters Shattered Moon, they have to go through all these areas with huge machinery, lots of broken and deformed metals, burning equipment, electric sparks, and fire smoke. We, the audio team, agreed that we wanted it to sound more like a factory where it has lots of creaking and stress, with some machinery sounds here and there.

Then, we get to the second visit, and the area looks more high-tech and not as busy. It has more computer console beeps, more modern servo sounds, and less metal creaks.

…we wanted it to sound more like a factory where it has lots of creaking and stress, with some machinery sounds here and there.

Shattered Moon also has two big elevators across the entire level. One is when Cal first lands on Shattered Moon and he has to take the elevator to go down, and the second one is for leaving the laser tube and going up to the lab. We differentiated the elevator sounds and used them as an opportunity to transition and tell the player that the ambience has changed. The first one sounds more like a rusted elevator and the one going to the lab is intentionally made to be more of a modern sci-fi and clean-sounding elevator.

AB: One of my favorite sounds I was in charge of was the laser in the Shattered Moon lab. The sound itself is relatively simple. It’s split into 3 sections: the impact, the body, and the tails. The main body of the sound is a “reese” bass with a second one high-passed and aggressively layered on top. The rest of the body is some granularized source from my “Koboh Tech” library I made for other interactables to add some warbly and shimmery textures. The initial impact is some heavily processed explosions and a laser-y sound Tom Jaine made for the library. The tails are a combination of processed interior explosion tails and some other Koboh tech granularized sounds. This is just one of the many sounds that amounts to the full experience of the moon’s laser.

The implementation of the laser, however, ended up taking on a life of its own.

The implementation of the laser, however, ended up taking on a life of its own. The sounds play mostly along a segment formed by the start and end positions of the blueprint. Along the segment, the main laser charge and fire sound play on the closest point to the listener, keeping them always close and present. However, additional high-end sweetener sounds also play when close to the laser on either end of the main sound to add width to the laser. The charge-up sounds play 4000 units away from the main laser sound in the direction of the start of the laser making it feel like the charge is always further down the laser and a subtle signal to the player what direction they need to go.

When the laser is charging, to add some extra motion to the sound, five “particle” sounds are positioned randomly along the segment and move towards the start of the laser at various speeds. This gives a subtle feeling of the laser pulling in energy and power towards the start of the laser and adding some interesting motion to the sound that pans appropriately with the direction of the camera.

…five “particle” sounds are positioned randomly along the segment and move towards the start of the laser at various speeds.

I wanted to make the player really feel like this laser was incredibly powerful so I started finding ways to make the environment react to the laser. Using the UE4 integration of Wwise callbacks, I added markers in the audio files so I could send callbacks to blueprints I scattered around the level to trigger different sounds at various times to be perfectly in sync with the audio. When the charge sound starts, I trigger one-shot sounds of the floor rattling around Cal, when the laser starts a constant rumble plays, and when the tails “release” the rattle fades away.

Additionally, anywhere the floor gets close to the laser, after the tails fade away, I trigger some cooling ticks on the metal surfaces to imply the metal is expanding and contracting from the heat of the laser. I also positioned some reflection sounds above and around the player so that when the tails play, you hear some extra positional reflections to help sell the power of the laser.

When outside of the laser, a new segment…plays an occluding version of the laser sound blended with some metal rattling sounds.

During this section, you can get outside of the laser at a few points. When outside of the laser, a new segment is created along the wall of the laser tube that plays an occluding version of the laser sound blended with some metal rattling sounds. Combined with the laser diffracting around the corners of the spatial audio geometry, this created a pretty convincing effect of how the laser is shaking the entire area.

Lastly, when inside the lab itself, there is a new processed version of the laser to simulate the sound of the laser firing from the other room but resonating through the large space of the lab.

Did you make any changes/improvements to the weapons’ sounds (lightsabers and blasters) for Jedi: Survivor? If so, what did you change/add/improve, and why?

[Note: These answers are meant to reflect changes and improvements we’ve made to these weapons since the previous game, not the franchise as a whole.]

OC: I was super excited to get to work on the combat finishers in this game and build upon the really unique and in-your-face style that we established in the first game. Compared to Jedi: Fallen Order, there were a ton of new finishers added and they had the potential of happening more frequently in combat.

The first thing I added was a kind of cinematic lead-in that would play when an enemy would be highlighted when a combat finisher was available. This acted as a nice way to lead the listener into being ready for the more over-the-top cinematic style of the finishers.

…I took a very modular approach in order to allow for quicker iteration and better consistency across the many finishers in the game.

With the sounds of the finishers themselves, I took a very modular approach in order to allow for quicker iteration and better consistency across the many finishers in the game. Along with some help from the other sound designers, we created a small library of different whooshes, impacts, crazy tonal lead-ins, and risers; I would then tag these in the animations for each combat finisher. My favorites are a lot of the aggressive feedback tones that I mostly made using stacks of flangers with the feedback turned way up.

After all the learnings I had working on the lightsaber for Fallen Order, I had a ton of ideas on many different ways it could be improved, particularly in terms of mix and fidelity. With the many varied lightsaber stances added to Survivor, it also meant the lightsaber sounds had to work in many more contexts than before. Our biggest goal overall was just making sure lightsaber combat always felt satisfying and exciting while never getting fatiguing no matter how long you played for.

Our biggest goal overall was just making sure lightsaber combat always felt satisfying and exciting while never getting fatiguing…

The first big thing I wanted to address was that the sound of the lightsaber swings felt good in isolation but because of the nature of lightsaber sounds being very bass focused, the swings often got lost in combat or levels with louder ambiences. The first step was to slightly adjust the assets so unnecessary frequencies were more aggressively EQ’d out and the sounds were made more tonal and cleaner.

On the mix side, I added more adjustments based on combat state, so when you are out of combat, the bass on the swings is more prominent. During combat though, a high pass filter is added in and the swings become louder, letting them be heard a lot more clearly in more contexts without muddying up the mix.

As for the different stances, most of the differences were handled in Wwise. Dual stance has the off-hand swings pitched up. Double stance also has the second saber slightly pitched but less so and has a very slight phaser added to the idle. Blaster stance shares the swings with Single, and Crossguard stance has unique assets.

Crossguard stance was a unique challenge because we very much wanted it to sound a lot heavier and more intense than a normal lightsaber…

Crossguard stance was a unique challenge because we very much wanted it to sound a lot heavier and more intense than a normal lightsaber, but it couldn’t sound unstable or too evil like Kylo-Ren’s saber. The first thing I tried was pitching down the regular saber sounds in Wwise but it still left quite a bit to be desired. What I ended up doing was making new assets using the original saber I had done as a baseline, pitching those down, and then adding a lot of additional processing on top to maintain the fidelity and make the saber feel weightier and more powerful.

I also created much slower lightsaber swing sounds to accent the more exaggerated movements Cal does in Crossguard stance. Just updating the swings alone wasn’t really enough to get Crossguard stance to feel much heavier than Single stance, particularly on the special heavy attacks. I added distorted air whooshes that play on top of the lightsaber swings during the heavy attacks and that is what finally got those to feel as heavy as they look.

Aside from generally just polishing the assets, I made new versions of the hits to account for the stances…

The next thing to address was the lightsaber impacts, improving the clarity of whether they were blocks, successful hits, or a kill. Aside from generally just polishing the assets, I made new versions of the hits to account for the stances; lighter hits for Dual and Double, heavier hits for Crossguard. There are also separate sets of hits for flesh versus droid enemies just to add a bit more variety and flavor.

Kills are mostly just more intensely tonal versions of the strikes but they also have a 2D stereo transient sweetener that plays to make them feel more satisfying but also more clearly communicates you successfully killed your target.

There are different intensities of kill sounds depending on the class of enemy, so when killing smaller enemies the kill isn’t much more intense than the normal hits, for example. But when killing a mid-boss type of enemy, the kill is significantly louder and more intense to add to the satisfaction.

I made different levels of intensity to add more variety when blocking combos from enemies using different intensities of attacks.

For blocks, I tried to make them feel a bit more rounded, as if the saber was hitting but failing to slice through something. I made different levels of intensity to add more variety when blocking combos from enemies using different intensities of attacks. The parry asset from Jedi: Fallen Order was used as a base for the new asset and I made three different variations: a “neutral” parry for when you parry a target but they are not staggered, a “success” parry for when the target is staggered by the parry, and then finally a more over-the-top parry for the Dual stance auto-parry special ability.

Instead of using HDR, in order to make sure a successful parry is audible no matter what the context, the parry events all have a global RTPC that ducks almost every other sound in the game for a fraction of a second; aside from helping with clarity, it adds a cool dramatic effect to go along with the slight slow-mo that happens when you parry.

…most blasters in the game now utilize a switch container to crossfade between close and distant variations of the main blaster sound.

AB: Early on in development, one of my first tasks was to build on the blaster and explosions system from the first game. One major difference in Jedi: Survivor versus Jedi: Fallen Order was the number of large open spaces as well as the ambient fighting that different factions like the Empire and Bedlam raiders would get into. This means that often you can hear blaster sounds from quite far away. To compensate, most blasters in the game now utilize a switch container to crossfade between close and distant variations of the main blaster sound.

The blaster was further broken down into parts by extracting the tails and reflections sounds and making that its own layer. Doing this allows us to control the sound of the reflections that are played based on the environment. Originally, there was also an additional low-end mechanical and bass layer that we played on top of the main blaster sound but that was combined into one layer for optimization reasons as we found simply too many enemies could be fighting at one time resulting in a lot of extra sounds playing.

When an explosion triggers, we cast 3-6 rays around the impact location and place surface-dependent debris sounds at those locations.

Explosions got additional love as well and, combined with our audio programmer Chris Hopkins, we made a debris and shrapnel system. When an explosion triggers, we cast 3-6 rays around the impact location and place surface-dependent debris sounds at those locations. This creates more realistic debris that allows for multiple surfaces and different widths depending on size. We can also trigger different sized sounds in addition to more locations to further increase the size.

A subtle addition is the random chance for a shrapnel effect to play as well. This means that up to two rays get fired towards either side of the player and a sound gets played and moved along that ray. This lets us play a sound of incoming shrapnel that whizzes past the camera and adds an extra layer of intensity to certain explosion sounds.

What were the biggest creative challenges you had in terms of sound on Jedi: Survivor? And how did you handle them?

NVK: This is now the third Star Wars game I’ve worked on, so I’ve become pretty comfortable with the franchise. I actually worry it would be harder to adjust to not working on a Star Wars game since it’s become a huge part of my design language. At first, when working on Star Wars, it can be very intimidating since it’s such a major part of our culture and the sounds are so iconic. But it’s so much fun to work on, and you actually have a ton of freedom within the universe thanks to our productive collaboration with Lucasfilm Games.

Pretty much every sound we add to the game from the films is modified to fit our style and mix.

The biggest creative feat we accomplished was just piecing together an overall vision for the sound. There are so many people involved in the audio for the game, and we want them to design not only in the style of Star Wars, but more specifically the Jedi style of Star Wars. Pretty much every sound we add to the game from the films is modified to fit our style and mix. And with so many talented people on the project, our style is constantly evolving throughout development. A lot of the AI sounds that were done earlier on in the project had to be redone to fit by the time we shipped.

Other than that, the biggest challenge was just the sheer amount of content we had to get done. I’m not sure if that’s a creative challenge necessarily, but we certainly had to get creative in order to pull it off! I want to thank our amazing audio producer, Crystal Lau. There is no way we could have shipped this game without her.

What were your biggest technical challenges in terms of sound on Jedi: Survivor? And how did you solve them?

NVK: When working on Koboh, there was tons of lush foliage throughout the planet. On Jedi: Fallen Order, we had a system for playing foliage sounds on Cal as he moved through the foliage, but we didn’t have a good way to play sounds on the foliage itself. I worked on a lot of the levels with foliage in them, and I remember also having a hard time getting the nice high-end detail you want in the beds in a way that felt realistic with what you were seeing on screen.

The various loops in Wwise are then affected by an RTPC for the number of plants and also the wind intensity.

I worked with Ryan Redetzke, our audio programmer, to develop a system for playing sounds on the foliage which responded to the type of plant and also the amount of wind in the area. Our system essentially looks for groups of plants of the same type and plays a sound on a multi-positional emitter that is attached to that group. The various loops in Wwise are then affected by an RTPC for the number of plants and also the wind intensity. It sounds like a minor thing, but it really adds a lot to the feel of the game, especially on Koboh. I also think it saved time with making ambience beds, because once I created the foliage sounds with nice high-end detail, we didn’t have to constantly add them into our beds and could just focus on the mid and low-end when creating beds.

AB: One challenge we had was getting realistic breath sounds for our creature AI. Oftentimes, breath would overlap with creature vocals or it would be interrupted in awkward ways and I wanted to find a way to handle this in a more elegant way.

We have a lot of creatures so the first challenge was making a breathing system that we could add to multiple AI and have it simply just work. The solution to this was making a modular blueprint component that we could add to our enemy blueprints and leverage shared logic.

We have a lot of creatures so the first challenge was making a breathing system that we could add to multiple AI and have it simply just work.

The component is split into a few sections: the first part handles the breathing cycle, the second part handles interruptions of the breathing system, the third part handles optimization and making sure the system isn’t running when it’s not needed, and the fourth handles inputs from the animation notify we use to modify the AI’s heart rate. There’s also an optional function that drives any animation sync in the few cases we sync additive animations to the audio.

The breathing cycle section is a series of functions that play our breath sounds and uses Wwise callbacks to avoid overlapping audio. When a play breath sound is triggered, it checks if it’s even allowed to play and if so, we lower the heart rate by a small amount and then post event callbacks and check for three things: the marker embedded in the file, the end of the event, and the duration of the audio file. The end of the event triggers the next function that adds a small pause between audio that gets smaller and smaller depending on how high the heart rate value is. The marker callback is a way to make sure the system doesn’t end up flipping on itself. The breath function works for both inhale and exhale breaths and this function tests a marker embedded in the .wav file that has either “Inhale” or “Exhale.” It tests to make sure that the blueprint and the .wav file are in sync. If not, it resets the blueprint to be in sync so we don’t end up with out-of-sync breaths. The duration callback simply keeps track of the time a breath takes to finish so that we can pass accurate values to the animation blueprint if we are syncing animations to the blueprint.

The breathing cycle section is a series of functions that play our breath sounds and uses Wwise callbacks to avoid overlapping audio.

The interruption section stops the breathing system from continuing and quickly fades out the breath sound. This interruption can come from one of two sources: a notify state we can place on an animation to add start and stop points for when we want the system to pause itself, or from our default AK Event notify that stops the breath for the duration of the event played on that notify. This way we can ensure that no breath audio is played on top of a creature’s roar or effort and that it restarts from the correct time. We also often use this on creatures we control additive animations for that we want to stop controlling the animation and have a bespoke idle animation and bespoke idle vocal sounds.

…I was able to feed a 0-1 float value to the animation team that they could use to drive blend spaces that stayed in time with the breath.

The optimization section is simply a periodic check to make sure that we aren’t playing audio when inappropriate. If the creature is out of range, we stop the system and restart it the next time it’s in range. If the creature is dead, we permanently shut the system down.

The last section is how we drive some animations. We realized that on some of the larger creatures in the game we could leverage this system to increase immersion. Using the duration of the breath sounds and knowing which type of breath was playing, I was able to feed a 0-1 float value to the animation team that they could use to drive blend spaces that stayed in time with the breath. On an inhale, the value lerps from 0 to 1, and on an exhale it does the reverse. We used this on the Spamel as well as the Mogu to drive chest expansion animations that play on top of the rest of the creatures’ animations.

Did you do any custom recording for the game? If so, where did you go, what did you capture, and how did you capture it? How was it used in-game?

AB: For most of pre-production, we were in quarantine so a lot of the time we had to record in our homes. This led to a limit on the kinds of things we could record. That being said, we found ways to find things to record. For the rushers (aka, Rawka), I bowed a plastic card and used the ability to warp the curve of the card to create formant filter-like movements and screeches that ultimately became about 95% of the Rawka vocals. I was able to also get different types of screeches by tightening the strings a ton and bouncing the bow off of the card to get stuttering sounds.

For the rushers (aka, Rawka), I bowed a plastic card and used the ability to warp the curve of the card to create formant filter-like movements and screeches…

I also bought some slime and a few random things like kazoos and trumpet mouthpieces and blew air into the slime to create garbled squeals and other wet slime sounds. This ended up in a lot of the Mogu vocals as sweeteners and “saliva” sounds to make it sound a little more gross.

OC: Most of the recording I did was just inside my room at the office, using my voice or various instruments and props I had laying around or went out and bought. The Spamel is almost entirely my voice with occasional layering of other animals to add more weight or interest. There hasn’t been much precedent in Star Wars for how a Spamel is supposed to sound so it gave me a lot of freedom to explore it. In order to make the Spamel feel as natural as possible sound-wise, I watched lots of videos online of camels vocalizing and then tried to imitate them as much as I could. By the very nature of using my voice, it sounded just off enough that it sounded a bit alien and different to a camel but still had a similar character to it.

By the very nature of using my voice, it sounded just off enough that it sounded a bit alien and different to a camel…

Another creature I mostly did using custom recordings was the Hardshell. I wanted to force myself to try and make something a bit more unique than the more common hissy type of insect sounds we’d usually do for this kind of creature, so I tried drawing inspiration from mole crickets and imagined the Hardshell producing sound in a similar way by stridulating using its mouth parts. This one was mostly a combination of castanets, a vibraslap, and my mouth sounds.

Rayvis is one enemy I did a lot of custom recording for. For his flail cable I recorded myself spinning a leather dog leash for some whippy kinds of sounds. I also recorded a lot of different sounds from my bass guitar, bending and pulling at the strings and tuning them up and down. The bass guitar gave me lots of nice tones I could use to make the swinging a lot more interesting than just an air whoosh. To add more cool metallic tones, I also recorded myself sliding two knives together (at home, of course) and then used aggressive denoising to extract a purer tonality from them. For both Rayvis and Caij’s blaster, I made all the source for them using the synth MeldaProduction’s MSoundfactory.

The last thing I did a lot of recording for was the Trident-class assault ship on Jedha. Most of its mechanical sounds and movements come from a vintage receipt printer I picked up at a flea market.

…the moment I saw the Relters on Koboh, I knew they had to be my cat. She makes the strangest little hums and chirps…

AF: It’s kind of a silly one, but the moment I saw the Relters on Koboh, I knew they had to be my cat. She makes the strangest little hums and chirps, so I’ve been waiting for ages to find the right spot to get her into a project. It took a few days of bribing her with treats and sticking a mic in her face, but I eventually got enough content to work with. Outside of one bird whistle recording I used to add some additional high-end, they’re basically all her!

Can you talk about your use of spatial audio in the game? What have been some of the challenges of working with spatial audio for Jedi: Survivor? What aspects of the sound work best in spatial audio? (Or, are there certain levels or fights that benefit the most from spatial audio?)

AF: On Jedha, there’s a sequence where the player is trapped inside of a sandstorm. While the beds for this moment came together pretty quickly, I felt that they lacked a sense of materiality from the player’s surroundings, as the debris collided with the props littering the field. To solve this, I created various loops of sand and rocks rapidly bouncing off of various materials like stone, metal, grass, and even plastic (for the Stormtroopers trapped inside with you). By attaching these loops to the correct objects and subtly modulating the pitch and volume of each instance, we ended up with a soundscape that subtly evolves as the player progresses through the storm and really shows off the power of 3D audio when it comes to ambiences.

Spatial audio pretty much enhances everything, but in a way that the player generally won’t notice.

NVK: Spatial audio pretty much enhances everything, but in a way that the player generally won’t notice. If it’s done well, things just feel “correct.” It was a lot of work setting it up, but with all the workarounds you have to do without spatial audio it actually ends up being the same amount of work if not less. Not saying it wasn’t a ton of work still; the game is huge. Shout out to Tori Ano, she did an incredible job setting up the majority of the spatial audio in our levels.

Jedi: Survivor used Unreal Engine 4. Was this a good fit for the sound team? You mentioned using Wwise middleware. Would you have been able to achieve everything with UE4’s audio system?

NVK: We did use Wwise middleware for the audio in the game, but UE4 was great for audio just because it makes it so easy for us to hook up sounds using blueprint. If we had to make code requests every time we wanted to add audio to game logic, it would really slow us down. So in that respect, UE4 has been great.

What were some things you did on the sound side to improve accessibility?

AB: I had the pleasure of taking point on the accessibility features of Jedi: Survivor. We had a number of audio features I’m quite proud of. The most obvious is the audio ping system. This is both an active and a passive ping system that alerts the player to interactables near them. The pings change sound by what type of interaction is nearby and more subtly change based on proximity and whether or not the player is in range to interact with the POI or not.

The pings change sound by what type of interaction is nearby and more subtly change based on proximity…

Additionally, we added audio ping support to the navigation assist tool. This tool places markers along the map that further granularize the direction the player needs to go to make it to the next objective. The navigation assist pings play sounds in the direction of the next location they need to go to and when the player reaches that point and the map icon moves to the next location, the ping plays a new sound at the next point letting the player know they’ve made it to the next spot.

…this is in addition to our intent to make the combat of the game as audibly readable as possible.

In the accessibility menu, we also added the ability to put the game into mono as well as turn down the ambiance from the rest of the SFX bus.

All of this is in addition to our intent to make the combat of the game as audibly readable as possible. Often during the development of our enemies, I would close my eyes and see if I could fight them with just audio alone. If not, that usually meant the enemy needed more work and polish.

What are you most proud of in terms of your sound work on Jedi: Survivor?

AB: I’m personally most proud of how the level ambiance and level interactables came out in the finished game. I ended up working on a large number of levels and it was a really rewarding challenge to make them feel unique but flow naturally from one area to the next. Between the systemic puzzle elements of Koboh like the bridges and orbs to the high-intensity areas like the laser on Shattered Moon to the quiet, decaying nature of Dredger Gorge, it was really fun to create compelling sounding areas that pushed the narrative forward.

OC: I’m really proud of how the combat sound turned out in general. It was my main focus throughout the project and it’s been really cool to see positive reactions to it in reviews and player reactions.

NVK: This was my first project in a lead role, so perhaps the work I’m most proud of is all of the things I did outside of actually working on the sound design. However, I did find time to do a lot of the UI SFX along with help from Tom Jaine and Steve Cowell (of Lucid Games). I built on the aesthetic I established on Jedi: Fallen Order, but incorporated some new palettes as well. My favorite thing to work on was all the popup sounds. Usually, it’s just some text and a line animation, but that gives you so much freedom to do whatever you want. I also love the challenge of just making a single sound that the player will hear over and over again, and being confident that the sound works without any variations.

…I just want to mention how proud I am of the team and what we accomplished!

Since this is the last question, I just want to mention how proud I am of the team and what we accomplished! It was not an easy feat, but everyone came together and really did some incredible work. I think we all feel super fortunate to work on this incredible project, and also with such a great team outside of audio as well. We are constantly inspired by the work of the Jedi team, without them we wouldn’t have anything to add our sounds to.

I’d also like to say thanks to the team at Lucasfilm Games. They were super supportive and gave us great feedback along the way to make sure we honored the timeless audio of Star Wars.

And finally, thanks for letting us ramble a bit on your website; it’s been fun!

A big thanks to Nick von Kaenel, Alex Barnhart, Oscar Coen, Tom Jaine, Ashton Faydenko, Kartika Luky, and Nicholas Friedemann for giving us a behind-the-scenes look at the sound of Star Wars Jedi: Survivor and to Jennifer Walden for the interview!

Please share this: