Moonspire on How Its Speechbound RPG Helps Learn Languages in Fun Way

Introduction

My name is Manuel, the CEO of Moonspire Games, a polyglot fluent in over five languages and the leading mind behind Speechbound – A Language RPG. Alongside Philipp Amelung, I co-founded Moonspire Games in November 2022 after working for over a decade in the visual effects industry as an award-winning visual effects supervisor. 

I first met Phil during our studies in Munich. I secured him a role at Prime Focus VFX in Vancouver when I was working as an FX Technical Director on Final Destination 5 in 2010. Despite Phil returning to Germany after the project, we maintained our friendship and professional relationship. 

Over the years, I've worked on four continents for some of the biggest visual effects studios, such as Weta, Industrial Light & Magic, and Sony Imageworks, contributing to major film projects such as The Hobbit 3, Avengers 2, and Captain America 2. 

In 2017, I co-founded Stormborn Studios, a company specializing in visual effects simulations. Stormborn quickly established a reputation for high-quality water and destruction sequences for clients like Warner Brothers, Apple TV, ABC, Amazon Prime, and Netflix. I left Stormborn Studios at its peak, having won several awards and an Emmy nomination for our work on the show 5 Days At Memorial. My passion for language learning and gaming led me to venture into a new challenge with Moonspire Games. Our mission is to revolutionize the possibilities of language learning through immersive gaming experiences, merging education and entertainment in a way that hasn’t been achieved yet. 

Moonspire Games

My journey towards launching Moonspire Games has been deeply influenced by a decade-long personal pursuit of self-improvement and language acquisition. I developed fluency in French, Spanish, and Portuguese while maintaining a full-time job, simultaneously studying finance and programming. 

Full immersion was the most effective way for me to learn a new language quickly. This realization led to the concept of Speechbound in 2021, which centered around creating a virtual environment where individuals are encouraged to learn a new language organically. The vision was to replicate the immersion experience of traveling abroad from the comfort of home. 

The primary goal was to ensure that the learning process didn't feel tedious, allowing for a natural progression and plenty of success moments to keep the player motivated. I wanted players to explore new worlds, engage in battles, and build relationships. Essentially, I was describing an RPG, but with the unique twist of language learning replacing traditional leveling-up mechanics.

Recognizing this potential in a rapidly growing market of gamified language learning, I decided it was time to lean into the adventure! I reached out to my former colleagues, Philipp Amelung (Art Director), Thom Holland (Creative Lead, Script Writer), and Jared Barber (Concept Artist, Animator), to form the initial core team. We spent a year developing the game's underlying concept, learning Unity/C#, and creating the first proofs-of-concept. We then expanded to a team of 12 within the next six months! 

Our team at Moonspire Games is now a mix of individuals from around the globe, collectively fluent in more than eight languages. The team's strength lies not only in its technical capabilities to create visually stunning and engaging gaming experiences but also in our shared passion for language learning and gaming. We have a team of seasoned programmers, artists, and writers, all committed to this big idea of changing the way that people learn languages. This diversity and dedication give us the ability to cater to the needs of a global audience.

Speechbound

The main difference between Speechbound and a traditional RPG lies in the fusion of language learning into the game's core mechanics and progression systems. While traditional RPGs revolve around character skill improvements and combat mechanics, Speechbound integrates language learning into its game design, making it a pivotal part of player progression. 

One unique feature of Speechbound is our advanced use of TTS (Text-To-Speech) technology. Traditional RPGs typically use recorded dialogues from voice actors, but because Speechbound is incorporating a whole dictionary’s worth of vocabulary into its mechanics, that’s not exactly feasible! Speechbound uses modern TTS systems to generate realistic dialogues in multiple languages, which not only enhances the immersive experience but also enables us to support a variety of languages adaptively. 

Another groundbreaking feature is our sophisticated language database. We have populated this database with a vast array of words, sentences, and their translations, as well as metadata in multiple languages. This allows for individualized user-score tracking and the generation of a multitude of dynamic language exercises. 

Our custom-designed dialogue system is another key differentiator. Conversations are a staple of RPGs, and Speechbound is no different! However, we’ve been working to create a unique system that incorporates aspects of learning into the conversations themselves. In order to progress in certain dialogues, the player completes language exercises to form, translate, or complete sentences. Acquired vocabulary during level exploration is reiterated, providing opportunities for better memory retention. 

In addition, to further reinforce long-term language acquisition, we utilize a custom Spaced Repetition Algorithm. This scientific approach to learning helps players review and reinforce their vocabulary based on their performance and recall rate. The algorithm we use intelligently adjusts to the player's performance, prioritizing repeated exposure to exercises that proved challenging, while less frequently revisiting those the player mastered with ease. This means that boss fights can really lean into testing what a player has learned in a given area! 

In essence, Speechbound marries the engaging, immersive world of adventure with modern, innovative approaches to language learning, augmented by cutting-edge technology in new and exciting ways. This blend is what sets Speechbound apart from both traditional RPGs and other language-learning applications.

Languages

At the heart of our game is a versatile database that supports language pairs. For example, our original prototype was designed around the English/Spanish pairing, offering players the option to choose either as their source or target language for learning. This means you can start with either language as your base and have the other taught to you! Our next milestone is the implementation of the English/Chinese language pair. Our robust database system can accommodate even the most complex languages, including those with Asian or Cyrillic scripts. 

Our ambitious plan is to launch Speechbound in Q4, 2025, complete with at least four language pairs, with English/French being our fourth target. Depending on the outcome of our Kickstarter campaign, we hope to support up to eight major language pairs at launch, with Portuguese, German, Russian, and Italian in our sights. The more funding we secure, the more expansive our language offerings can be.

Visual Development

We wanted to create a unique visual language that distinguishes itself from its peers. Set on the sci-fi world of Babyl Moon, with a sprawling city at its core, our design style features vibrant, saturated colors, Solarpunk-inspired architecture, and cute anime-style chibi characters. Initially, we planned to build our world as a 2D tileset, but due to terrain complexity and the challenge of having different level heights that the player can traverse made us reconsider our approach. Now our world is built in 3D, with a top-down camera that shows a slight parallax to make it extra pleasing to explore. We spent more than a year in look development in order to establish our current aesthetic, which we’re very happy with as it continues to evolve. 

First, our amazing 2D concept artists Zhiting and David as well as our Art Director Phil do detailed artwork that depicts the environment. In the first step, we define basic shape language, overall design, scale, and shapes. Once happy with the initial design, we add more detail and establish color palettes. 

Once the art department is happy with the look, we start building assets in 3D using Blender. Phil usually translates the 2D images to 3D with the help of Stefan, our class-A 3D Generalist. The next step is to make sure the topology is optimized for video games. Additionally, we keep our assets as modular as possible in order to be able to create a variety of different assets with just a few base objects. This also allows us to take advantage of instancing (prefabs) in Unity and quickly block out environments. 

By strategically using a palette mapping technique for the base colors, we keep texture load and draw calls as low as possible. We are controlling the color assignment and overall look with the powerful in-house Speechbound principal shader our art department created. This shader also has a couple of extra channels for vertex color information, which enables us to bake vertex-AO (Ambient Occlusion) directly into our assets without the need for baked texture maps, and also allows us to apply additional vertex colors to further break up the base layer. For additional detailing, we use decals and trim sheets. 

Our game engine of choice is Unity – it’s just so flexible and accessible for a team to really dig into creatively! We’re using the Universal Render Pipeline (URP) to render our game, including 3D assets and VFX. The URP allows us to create high-quality visuals both for our desktop and mobile users, without sacrificing performance or compromising on looks. Real-time colorful illuminations, dynamic shadows, HDR, and lots of tools are at our disposal to create some really cool visuals. We’re also utilizing custom shaders created specifically for some sweet unique looks! Unity enables us to use state-of-the-art tools and technologies to further improve the user experience on whatever platform they play.

Audience

I'm often asked, "Why take on the added challenge of incorporating language learning into an RPG? Isn't the creation of an RPG already complex enough?" The answer lies in our desire to carve out a unique space within the gaming landscape. Yes, developing a new form of language learning experience is indeed a tremendous undertaking that demands extensive resources and intellectual effort. Yet, if we can successfully realize our vision of an immersive, visually stunning, and gripping gaming experience, then Speechbound will stand as a shining beacon amidst a vast ocean. We aim not merely to exist among the multitude but to illuminate the path to a unique and exciting learning experience. 

Finding an audience for an educational game like ours presents a unique challenge. We've initiated a comprehensive social media campaign across multiple platforms, with a special focus on TikTok and YouTube, since these allow us broad reach and a lot of diversity in what we can show our audience to really get them fired up! Alongside this, we just established a dedicated Discord server where we've started building a vibrant community of language learners, encouraging mutual growth and a shared passion for our game. We hope to foster a really inclusive and warm environment for our fans and players there. 

The landscape of game development in 2023 is complex. On one hand, social media offers an unprecedented opportunity to reach gaming and language enthusiasts worldwide. On the other, it demands substantial resources to maintain momentum and audience engagement. To stay ahead of this, we're producing a range of content, from behind-the-scenes footage to painting time-lapses and language-learning videos. This not only keeps our creativity flowing but also gives our audience a peek into our world throughout development. 

As an independent studio, we count on our community's support. We're planning a major Kickstarter crowdfunding campaign in Q1 of 2024 to bring Speechbound to as many players as possible. We truly believe in our vision and are doing everything we can to realize it. We hope that you, our fans, recognize and support our efforts. You can help us by engaging with our content on social media - every like, share, and comment truly makes a difference. Upon our initial release, Speechbound will be available on Steam, iOS, and Android platforms. We then plan to extend our reach to Playstation, Xbox, and Nintendo Switch. 

For the latest updates on our Kickstarter and other important news, please sign up for our newsletter on our website. Your support is crucial to our success, and we appreciate it deeply. Together, let's revolutionize language learning through gaming!

Manuel Tausch, CEO of Moonspire Games

Interview conducted by Arti Burton