Making the magical sound of ‘Ori and The Will of the Wisps’

Moon Studios’ release of Ori and the Will of the Wisps has been one bright spot in bleak March, delivering a sorely-needed dose of ‘awww.’ Its predecessor,Ori and the Blind Forest, was a singular game experience — an IP that’s instantly identifiable and universally loved, thanks to the gorgeous visuals, beautiful story, evocative score, and engaging sound. But if Blind Forest was a rainbow-sprinkled donut then Will of the Wisps would be a dozen of them. It’s the same adorable Ori but with many new abilities, new weapons, new locations, and new boss encounters. And Ori’s story progresses like you wouldn’t believe (I won’t ruin it for you!).

Formosa Interactive took on the role of the audio department for Moon Studios, handling everything from cinematics to implementation. Here, Audio Lead Guy Whitmore, Audio Lead Alexander Leeman Johnson, and Senior Sound Supervisor Kristoffer Larson talk about the opportunities and challenges of furthering Moon Studios’ creative vision and contributing something special to this game series.

Composer Gareth Coker’s score is so stunning — one of my favorite game scores of all time. How did the sound of his score influence your approach to sound effects on Will of the Wisps?

Kristoffer Larson (KL): From a title perspective, the score has always been a major component. The game itself is very musical and lyrical so that always needed to be at the forefront. We never wanted to get in the way of the score. The score is part of the Ori DNA and we took that into account.

Senior Sound Supervisor Kristoffer Larson

We had a lot of temp tracks to work with. We made sure there was good coordination between Gareth and our team, especially with me in terms of the sound effects and mix perspective. I wanted to understand the frequencies that he would be dominating and places where effects would dominate, and the times when music would be up front and where sound effects would be up front. So we took all of that into account when we first started.

Guy can speak more to the specifics of how he dealt with that on a day-to-day basis.

Guy Whitmore (GW): Gareth had some early tracks that he created. The desert and the swamp levels were pretty close to how they ended up without the live instruments. So we had a good flavor of where he was going; that really helped.

Audio Lead Guy Whitmore

In terms of ambience, I always thought of how it was subliminally supporting the music. I wanted to support the music most of the time, unless of course it was a moment when sound was forward in the mix. The goal for ambience was to find ways to complement the music. The game is very impressionistic, so in terms of ambience that is the approach we took. It’s not overly literal. It’s lush.

It was good to be in touch with Gareth quite a lot. He even shared with us some of his processing techniques, like reverbs in particular, so we could mirror those in the ambience, so the color and types of sounds would be complementary to one another. At times, it would be very literal, like here’s the reverb I’m using and other times it would be an explanation of the type of sound he was going for.

Once the live instruments were added, the dynamics and treatment felt great. It became easier for all of us to fit what we were doing around the music as those mixed tracks came into the game. I could really move at a quick clip and know that the music tracks would be dropping into place.

 
Did Formosa Interactive handle the in-game sounds as well as the cinematics?

KL: That’s one aspect we’re really proud of, that we were essentially the entire audio department from conception to execution. We were an entirely outsourced audio department. That works well because Moon Studios is decentralized; they don’t have a common facility. All of the developers are working from around the world. That worked quite well for us. We have established a process that allows us to be a team that can drop in to help fix specific content issues or tech issues, or take on entire games as an external sound department.

Since the game was created with the Unity Engine, were you using Wwise or FMOD for implementation?

KL: We ended up using Wwise and that was a choice that was already made ahead of time. We have a significant amount of experience with Wwise, as well as other audio engines.

Using Unity and Wwise is something the team was familiar with but the caveat there is that every developer — whether they’re working with Unity or Unreal — has their own way of using the engine, like they do timelines their own way or animations in a particular way or VFX in a specific way. So knowing Unity is a great place to start, but when working with developers who are very adept at modifying engines to suit their needs, the implementation becomes quite unique to that situation.

What we’re able to do is jump in and learn a developer’s method and process and apply our experience and expertise so that it’s not completely new; it’s only a flavor of new.

What we’re able to do is jump in and learn a developer’s method and process and apply our experience and expertise so that it’s not completely new; it’s only a flavor of new.

One thing that did help was because Moon Studios is decentralized, they do a lot more communication via videoconference. They create videos and are very much online so if any of us had a problem or an issue or question we could hit up a developer on Skype and they would make a five-minute video of how to do something or we’ll do a screen share. They can show us how they do something and it’s very quick. It’s pretty much the digital version of walking over to someone’s desk and looking over their shoulder as they show you how to do something.

There was a lot of learning on the spot. Moon Studios develops tools very quickly and very reactively to the needs as they arise. Like a lot of developers, as soon as something is documented it is already old. So they do lots of videos explaining how things work on the day. Those were really helpful and crucial for us, to help us figure things out in terms of their process. But we’re very good troubleshooters and good communicators, and that allows us to be as productive as possible.

Looking at the sound design side, were you able to use any assets from Ori and the Blind Forest, like the UI/UX sounds, the reoccurring abilities like double jump, and some of the damage sounds?

KL: Yes, definitely. The intention was always to bring Ori to the modern platforms/consoles and do an up-res version. We didn’t want to take the amazing game that Blind Forest was and throw that out. We didn’t want to start anew. We wanted to be preferential and reverential to the original core aesthetic and DNA.

There were essential elements that we wanted to keep and had to keep, like Ori’s movement and the UI that players are familiar and comfortable with. Our goal was to bring that to this modern version. It was like taking some sounds and remastering them.

Alex did a big chunk of that because most of the sounds that were not new were for Ori’s locomotion and UI.

Alexander Leeman Johnson (AJ): There were a lot of iconic sounds in that first game, and I’m coming at this from the perspective of a big fan of the first game. I played it, I loved it, and I love how it sounded. Nearly five years later, it’s crazy to work on the sequel and be able to lend my hand and my ear to it.

Audio Lead Alexander Leeman Johnson

There are a lot of iconic sounds from the first game that you can’t fundamentally change, like the jumping sound. There’s this breathy, springy texture to it that’s actually Ori’s voice and so you can’t mess with that.

In the first game, I don’t think they had enough horsepower for dynamic reverbs. So a lot of the sounds had reverb baked into them. One thing we did early on was to go through original sessions from the first game and pump out those assets fresh, without the reverb baked in. There were environmental reverbs and a special bus just for Ori. So, all of Ori’s sounds we could pump through this bus and get a special Ori reverb that glues the character together in real-time.

On top of that, there were a lot of new sounds because Ori has new abilities. In the first game, there was the Spirit Light attack ability, Stomp, and Bash. But now they’ve added another 10 or 12 things you can do.

And also the damage is scaling, so you can upgrade these weapons over time. You can stack different abilities using the chart system. There’s a new bow that at first is just a normal bow and then you get the Splinter ability and it splits into three. You can upgrade the Splinter ability so it splits into five. The arrows are bouncing off enemies that have armor. So we had to create these new sounds from the ground up that would coexist in the fabric of legacy sounds we wanted to bring from the first game.

Was it difficult to match the tone of the original sounds, to create something new that had that same Ori-ness to it?

AJ: At first it was kind of tough. I wanted to do right by the game because I’m a real fan! By the end, I really got a handle on what needed to go in — lightning, thunder, ice, the right kind of tonal elements, and whooshes. I could build a weapon from the ground up pretty quickly by the end.

What went into creating the tonal elements? Was it soft synths?

AJ: I used a waterphone. I bought a few sound libraries that have some of that in it, like Tension . Waterphones have a sonically rich tonal element that you can put in there.

On the grenades, I also ended up using gasps and breaths, which have their own tonal characteristics.

I tried to use sounds that were organic or had a performative element to it. That was the key to making sounds for Ori’s abilities.

I tried to use sounds that were organic or had a performative element to it. That was the key to making sounds for Ori’s abilities.

KL: That was one of the biggest challenges because Ori is a being of magic and light, of energy and spirit. His world is a very fantastical, magical place. But trying to make that believable and organic and not have it sound sci-fi or synthetic or artificial was the real challenge. Especially when you’re talking about things like energy and weapons, you can’t necessarily grab Massive or your Moog synth and pump out some tones. You have to think a lot more about how you create sounds that have natural believability — it’s magical but isn’t windchimes, it’s musical but doesn’t sound “Buck Rogers” or super high-tech.

We ended up using a lot of natural source and using effects to modulate those, to add shimmer while still retaining a lot of the organic nature of the source material so that your ear picks up — even on a subconscious level — that this is a natural source that has been re-contextualized with bright sparkly effects. The source might have been ice or rushing water or something natural but when you use some modulation and couple that with some magical effects then the brain does not see it as just water or just ice. It fulfills that magical element while letting the listener believe what they are seeing. The listener isn’t able to identify that the source element was a volcano, let’s say.

Any particular plug-ins or effects that were helpful in modulating these natural source sounds?

KL: There wasn’t just one thing that we used over and over; there wasn’t an “Ori-verb” that we put on everything. Instead, I would decide what it was I wanted to achieve sonically, and then use the best tools to help me achieve that. If you smother “Ori-verb” on everything then it’s all going to sound samey-samey. Even if you do have some favorite go-to tools, you need to have variety, both in frequency and timing.

Sometimes I would use Valhalla Shimmer because I really like how well it disperses, but a little goes a long way. You can’t just use that all over the place. So I would use that on some things but I would also choose other tools to create a similar effect. For instance, I would use Eventide’s UltraReverb which has a lot of really great modulation. I would build a shimmery patch that might be similar but is going to sound different. You create lushness with variation.

There are a lot of tools that do similar things, but each has its own nuance of color and flavor that it lends to the sound.

There are a lot of tools that do similar things, but each has its own nuance of color and flavor that it lends to the sound. In this situation, you are creating this giant orchestra of sound and music and you need variety so the sounds will stick out in the mix. Otherwise, when mixing, you have to make room for the sounds because they all end up occupying the same frequencies. If you build in that variety while creating the sounds that helps in the end mix. Especially for games, I’m always looking at it from an end mix perspective.

In an ideal world, we would have at least two weeks to do a dedicated mix, but the reality is you end up with elements coming in late or bugs to fix. So the mix needs to come together as you are building the game, and so the more you build variety into every asset and every routing, and every implementation, the less work you have to do to try to wrangle the mix. It just creates itself organically.

What were some of your sources for sounds? Any particular libraries that were useful? Did you do a lot of custom recording?

GW: There were a lot of library sounds but, because this is such a nature-oriented game, I did record some foliage. My wife has a garden and I went out and recorded all different types of leaves and sticks and built up a library of foliage. Some of that I used for wind and for leaf movement.

For background critters, I bought a library of hunting calls — everything from moose calls to turkey calls to duck calls to crow calls. As Kristoffer pointed out, we wanted something that was otherworldly but not synthetic. So we used organic sounds to mimic creatures and that was a good way to go. Through processing, we could make those sounds evoke a feeling of otherworldliness.

…we used organic sounds to mimic creatures and that was a good way to go. Through processing, we could make those sounds evoke a feeling of otherworldliness.

KL: I’m always recording. I didn’t necessarily do any specific sessions or content recording for Will of the Wisps, but I did draw from my own recordings that I’m always collecting.

One of the benefits of Formosa Interactive is that we work on a lot of games, films, and series so we have a rather extensive collection of commercial libraries and bespoke content that has been recorded for other productions. We have deep library-pockets to pull from. That definitely helped a lot.

One thing I did request specifically from our librarian was to purchase the Bowed Cactus libraries from recordist Thomas Rex Beverly. There are two of them available. Those are really great libraries because the sounds have natural modulation but it’s not like an LFO. It has perfectly imperfect cycles in it. The recordings were captured at 192k, and so they pitch down nicely. When doing warbling creatures and textures, it’s a really great resource.

AJ: Absolutely! I used that quite a bit on some of the characters. I ended up focusing on a lot of the enemy design. So I used the Bowed Cactus libraries, my own voice, and the Animal Hyperrealism libraries. Those are really wonderful because some of the files are 32-Bit 384k recordings; they’re super clean. There are recordings of exotic birds and birds that I had never heard of, and those can be pitched down to 5% and they still sound clean. They don’t sound muffled. You’re essentially pulling them down 4 or 5 octaves. You can roll off the lows and end up with sounds you have never heard before. That ended up being quite a benefit.

Let’s look at some of those enemy sounds. Starting with Howl, how did you come up with its sound? What were some of the challenges in creating that battle encounter in the game?

AJ: It’s a really key moment in the game and a great place to start thinking about it. It lets you know that in this game there is going to be chases and boss battles and sequences where you have to execute the moves just right, or you die and have to start over. It’s a challenge that exists in this game and it’s what makes it fun!

So there’s this wolf that will insta-kill you if you don’t do the chase sequence properly and then you have this boss battle at the end of it. If you die there, then you have to start back at the chase sequence.

This is a big, key moment and it was important for the character to have the right amount of threat and weight and also some variety in the sounds that he makes. Even though the chase is fairly linear, if you die and start over a couple of times you want to have some dynamic quality to it. So I built him out as a fully-kitted character, with individual sounds that can be rearranged. If they had wanted to add another 30 seconds to the chase I would’ve had enough assets there to do that.

This is one of the places where there was a need to work between music and sound effects because everyone wanted to be there, all in the same frequency range.

The mix for this sequence was fairly challenging because we wanted the character to have dynamic impact but it’s also a very intense sequence. So there are moments of roaring and howling. We wanted it to have impact but not be too busy; that was the real challenge.

KL: This is one of the places where there was a need to work between music and sound effects because everyone wanted to be there, all in the same frequency range. The direction from the developers was for this encounter to be super big and dramatic and super impactful.

As we all know, if you push everything to 11 then nothing sounds big. It’s either just noise or you end up turning everything down and then it’s too quiet.

Ori and The Will of the Wisps composer Gareth Coker talks about creating the score for the game

This was a section of the mix where I needed to look at the frequency ranges that were being occupied. Gareth’s score had this great theme for when Howl comes in. There’s this choir and the music is pumping and it’s using all of the frequencies. And the choir was exactly where the ‘voice’ of Howl would sit. So I needed to do some surgical mixing, to find space for both the choir and for Howl’s voice.

AJ: We looked at a spectrograph of the music and picked out particular frequencies that would work for Howl, so we’d be able to hear them both. That’s something I did in the design phase and then handed it off to Kristoffer to mix.

KL: As far as content is concerned, Alex used a fair amount of animal source sounds. But the other directive that the developer had (and we knew we had to have as well) was that each character needed to have a signature set of sounds — its own believable and performative elements. We did a number of passes on most of the bosses in order to dial that in for each one. We did Howl three or four times. It was always pretty good but lacking in one thing or another.

We ended up eventually using a combination of animal source sound layered on top of voice actor Patrick Seitz’s performance of the sequence, which he did linearly. He has this great deep voice and he just wolfed it out. He did five or six passes and Alex used that as a skeletal structure of the personality for Howl, so he felt like a believable, tangible creature. It was all about getting that performative aspect from the voice actor and then decorating it and processing it in order to bring it into the nightmarish world of the game.

Let’s look at the Hornbug boss fight. This bug is certainly different from Howl in many ways. How did you approach the sound of this boss, and what were some of the challenges in creating that battle encounter?

AJ: That was an interesting boss fight because there is a lot that needs to be telegraphed clearly. This is a good example of what we were up against. This is a very natural, beautiful, organic game but that’s the presentation of this very mechanical Metroidvania game. At its core, it’s a very tight platformer with very tight things happening.

The Hornbug has a rhythm and a pace to it that is all about when these big, definite attacks are going to come. For example, when the Hornbug does its ground attack (not the fire attack, but the ground explosion where spikes come out of the ground), to really communicate that you need to balance the sound of the spikes so that they’re present enough. For the majority of the attacks from this character, I found that I could build in a sound cue before the attacks would happen — either seven frames or 14 frames in advance. So technically I was able to build that in and supplement the rhythm that the character already has. So as you are playing, you are being warned in a way that you can subliminally become comfortable with, if you were to fight this boss a couple times.

Another thing I did was to take bottlenose dolphin clicks from one of the Animal Hyperrealism libraries and pitch them way, way down. So you’ll hear these clicks as a warning before a really big attack.

KL: I remember doing some pulls for you of insect wings flapping. That’s one thing I felt it was missing looking at it from a mix perspective.

…so much of combat and enemy design is finding the balance between the aesthetics of what you are supposed to hear and feel, and user feedback.

As Alex was saying, so much of combat and enemy design is finding the balance between the aesthetics of what you are supposed to hear and feel, and user feedback. You need to know when there is an attack coming, if you hit the enemy, if the damage has been blocked or it was successful. So there were a lot of things we needed to balance looking at it from a system perspective. We needed rich combat that was informative and clues the player into what they should or should not be doing. The challenge is how to get the character aesthetic and personality into that.

A lot of what I pulled was this modulated, shuddery, stuttery insect wings and modulated clicking. It’s interesting how that’s coupled with a warbling, screechy vocalization. Those sounds really cut through the mix. It’s this giant creature and the camera is shaking all the time. We are trying to convey a sense of weight, so there is a lot of bass and low-mids that are in the mix that can get overwhelming really quickly, especially when we need a big roar and there is just no room for it.

Those clicky sounds gave it the feeling that this is a big, gross bug. We didn’t pitch the clicking down that much in order to convey the size because we already had a lot of that energy in the lows for his body movement and pounding and stomping. The amount of pitching that was needed to convey ‘giant insect’ was surprisingly minimal.

We look at all of the components going into the sound, but it’s important to look at the end product, to look at the fully baked cake. That is what the goal is. When you have a lot of sounds performing the functions you need them to do, you don’t need more of those. You just need a little bit of frosting on top in order to sell the sweetness of the cake.

Howl and the Hornbug are both animal/creature-based bosses. But Ori also fights a light-based enemy called the Willow Stone. How did you approach that boss fight?

KL: We dubbed that one the ‘Laser Shooter.’ This is a great reminder that this is a fantasy game. There is nothing techie or sci-fi about it whatsoever and anything that had a whiff of synthesizer would immediately get shot down.

So the Willow Stone has a laser beam and it telekinetically throws rocks. That was a challenge. We relied heavily on natural source. And luckily, all of his defenses were made out of rocks that he pulled from the environment so a lot of the movement was rock-based.

When I was working with one of the designers who made that boss, we needed to figure out how to make a laser beam that doesn’t sound like a laser beam. We didn’t want to get too complex. We went with a layered system that’s using natural, organic source and modifying and modulating it in a way that we get some irregularity.

…we needed to figure out how to make a laser beam that doesn’t sound like a laser beam.

One of the layers was rushing water but not so uniform that it was white noise. It had to have this liquidy texture and dynamic movement. We also used some ember sizzle and a bit of an earthquake rumble. But they were all these very natural sources that we layered in. We used a good amount of EQ to make sure it wasn’t a Dagwood sandwich of everything.

We had a specific function for each layer – this was the energy layer, this was the body layer, this was the burning layer, and so on — and used EQ to fit the sounds together so the sizzle was on top, the body was more low-mids, and the movement of the rushing water was obscured enough so that we could cue the movement without it sounding like water because it’s supposed to sound deadly. Using that combination of sounds and that approach allowed us to create an organic laser. And so that’s what we ended up doing for anything that had a beam.

AJ: There are a bunch of them. Both the spider boss and the owl boss at the end had a laser element. So we tried to nail these down to be something that is appropriate.

The spider boss had a lot of acidic stuff going on so I could incorporate that. For the owl, I ended up going with a more spiritual sort of sound, adding in elements that you would use for ghosts and so on.

Let’s look at some of the vocal design work. In the beginning, as they tell the story of Ku, there are vocals for Ori, Naru, Ku, and Gumo. They’re adorable! What were some of your favorite vocal designs throughout the game?

AJ: Kwolok was an interesting one because, like many of the characters in this game, he’s not one-dimensional. They have some back story on the reason they became bad. With Kwolok, he’s around in the first part of the game as a friendly character. He goes through a transformation when something horrible happens to him and you fight him as a boss. That was a really interesting thing to do, to have him as an NPC and then as an enemy.

KL: I did all of the VO direction for all the voices — whether it’s for the NPC Ori speech or gathering the source content for the design.

A few of the characters stuck out for me and it’s because of how well they landed. The first was Mora the spider, who was voiced by Tiana Camacho. In her audition, she did this amazing backwards speech that was 100% in her performance. She did this whispery, backwards-inhaley thing that sounded so cool. It wasn’t editing; it was just her. So we capitalized on that and that was the bulk of the ‘processing’ for her character. We did some modulation to give it a little tremolo and used EQ to make it more breathy and hissy. But all of that backwards feel was in her performance.

For the shard trader Twillen, the direction we wanted to go for him was almost alien. He’s this elusive character who doesn’t seem like he’s a local. He has this very odd delivery that, with a little bit of phase and flange, was delightfully unsettling and alien sounding without him actually being a creature. It’s almost like a creature in a person’s skin.

The Moki — the little otter-like creatures — are definitely a favorite among the developers and fans. They’re just cute — the way they are portrayed and animated, and the lines that were written for them. We wanted to do them justice. That’s a combination of me and Tiana. Both of us did a few passes of cute little chirpy-squirrely vocalizations.

All of that content was taken by Alan Rankin, one of our sound supervisors here at Formosa. He took that source and edited it together so that it felt cohesive. When we were recording the voice, we didn’t have the full breadth and scope of the game laid out yet and so we had to make a lot of toolkits in order to create content on-the-fly as vignettes and cut scenes came together, or as characters were introduced.

So those were voices that stood out from an NPC perspective.

In terms of sound, what are you most proud on Ori and the Will of the Wisps?

GW: I’m most proud that all the elements function well together and are complementary in character and to the essence of the world, especially given the short timeframe allotted to create, implement, and iterate these elements.

AJ: I’m proud of the work I did on Ori’s weapons and abilities, like the grenade, hammer, and swim dash. Ori is a unique character, like a spirit partially projected into the material plane. Because of this, there is a fairly strong design language that unifies the player abilities – elements like thunder, ice, wind, and organic electrical discharges dominate the palette. Designing clearly distinctive sounds with this in mind was a challenge that was very rewarding in the end.

KL: From a sound perspective I’m proud of the fact that we did exactly what we intended to do: take the DNA of the artfully crafted, award-wining original game and create a wholly new soundscape that is familiar yet upgraded and presented in Dolby Atmos! But I’m most proud of my team here at Formosa Interactive. Everyone pitched in on this one and, despite the many challenges, we helped put out another great sounding game.

A big thanks to Guy Whitmore, Alexander Leeman Johnson, and Kristoffer Larson for giving us a behind-the-scenes look at the sound of Ori and the Will of the Wisps and to Jennifer Walden for the interview!

Please share this: