Learning English can be challenging, whether you’re a beginner or aiming to improve. But did you know that there’s an app that helps you learn through popular manga like Demon Slayer and Kimi ni Todoke? We spoke with Takeshi, Kazuhiko, and Timothy (“Tim” for short) from Mantra, the company behind the popular Langaku app, to find out more about this exciting language learning tool.
Takeshi Yamanaka | Product Manager, Langaku, Mantra Inc.
Takeshi worked in the publishing industry and at an advertising agency specializing in games before joining Mantra in 2019. As the product manager for Langaku, he oversees user interviews, planning, and other tasks.
Kazuhiko Yasuda | Software Engineer, Mantra Inc.
After working on data analysis infrastructure development and operations at Yahoo Japan, Kazuhiko joined Mantra in 2021. As engineering manager for Langaku, he leads and drives development.
Timothy Lui | Product Designer, Mantra Inc.
Having worked as a graphic designer and in web app design, Tim joined Mantra in 2024 as the first in-house designer. He is responsible for UX design, as well as various other design tasks across the company.
Langaku: An App That Teaches English Through Popular Manga
First, can you tell us a bit about “Langaku”?
Takeshi: Langaku is an app that helps you learn English through popular manga, making studying English fun and accessible for everyone, whether you’re a beginner or an advanced learner.
The company behind Langaku, Mantra, initially focused on developing manga translation technology. By combining that technology with my own experiences, the idea of using manga to learn English came to me. I thought it was a unique and exciting approach—and that’s how Langaku was born.
At its core, Langaku is a learning app. We wanted to make the often challenging process of learning English more enjoyable by combining our technology with the engaging world of manga. Our goal is for users to naturally absorb the language, while having fun in the process. We hope to have a positive impact on learners’ lives by making language learning easier and more enjoyable.
Tim: I originally taught English as an ALT (Assistant Language Teacher), but I felt there were shortcomings in the way English is taught in Japan. When I learned about Langaku, I thought, “This is awesome!” Many people overseas learn Japanese through anime and manga, and I believe this approach can lead to great results.
Surprising Success: A Screen Blending Japanese and English
How do you arrange the manga translations?
Takeshi: The translations shown are typically the official English versions from American publishers. Since this is an English learning app, we wanted to ensure users are exposed to authentic English that’s also enjoyed by English-speaking audiences. That’s why we use the official English manga translations.
Meanwhile, we use AI technology for unique features such as switching from uppercase to lowercase letters and toggling between Japanese and English when users tap the screen.
The feature that lets you control how much English appears on the screen is really unique. How do you decide which panels stay in Japanese and which ones are switched to English?
Takeshi: We have four settings that allow users to control how much English is displayed: “Low” “Moderate,” “Heavy,” and “Full English. It’s actually pretty simple. Panels with more dialogue usually stay in Japanese, while panels with less dialogue are more likely to be in English. We also make sure the amount of English is balanced across each page to avoid overwhelming the user. If there’s too much English at once, it can be stressful, so we adjust things to help users ease into it, starting with shorter bits of dialogue.
Did you have any discussions about how to mix English and Japanese on each panel?
Kazuhiko: In earlier versions before the release, this feature wasn’t included. The display was entirely in English, and you could tap to switch to Japanese. However, since the first screen you see was all in English, users often felt overwhelmed, thinking “This is too difficult.” While discussing how to alleviate that pressure, we came up with the idea of displaying a mix of both languages. The thought was, “Since users are already tapping to see the Japanese, why not show some Japanese from the start?” After testing it, we got a good response, and that’s how we decided to implement it.
A glimpse into the background of English learning in Japan,based on user behavior
How do you capture the feelings of English learners, such as the sense of “overwhelm” or “pressure” that we mentioned earlier?
Takeshi: We regularly conduct user interviews, usually once a week, and always carry out usability tests after major changes. By observing how new users feel and interact with Langaku, we can adjust and fine-tune things based on their feedback, which then informs new features and improvements.
What kind of insights have you gained so far?
Takeshi: A recent realization was that new users often tap on every panel when they first start using the app. When we looked deeper into the reason for this behavior, we found that users were tapping to try to figure out the Japanese translation. It felt like a quiz where they were testing themselves. However, because many parts are paraphrased in the translation, even if they look at the Japanese, it doesn’t always feel like the right answer. Plus, it takes extra time to read, which leads to user frustration and stress.
Is there a way to reduce that stress?
Takeshi: This is something we discovered during the interviews as well. At first, users tapped on every panel, but over time, most of them stopped doing that. To encourage a more relaxed reading approach, we show tips such as, “It’s okay to skip parts” or “You don’t need to understand everything.” After seeing these tips, many users changed their behavior and adopted a more relaxed approach. However, a lot of users continue to skip the tips, resorting to figuring it out on their own. This is something we’d like to improve in the future.
Personally, I feel that the way people approach manga reading might be influenced by their traditional reading habits—such as not paying attention to every little detail in each panel.
Takeshi: I think part of that comes from the experience of traditional English learning. Historically, English education in Japan has emphasized translation from English to Japanese. This approach stems from the need to understand by translating English texts into Japanese, and English education has followed this tradition.
Currently, the focus is shifting more toward communication skills. The ability to understand authentic English in its natural setting is becoming more important than simply translating it into Japanese. Curricula have been evolving to reflect this, but for my generation and those before it, there was a strong focus on memorizing vocabulary and grammar, or on the skill of translating English into Japanese. As a result, many people still feel the urge to translate English into Japanese when they see English text.
Tim: On the other hand, this realization helped us understand the needs of the users. Initially, we only had a simple dictionary, but we realized users wanted more detailed information, including individual words and grammar. So, we’ve added an AI-powered dictionary function based on this insight.
Using OST to Organize Insights from User Interviews
After hearing from users, how do you incorporate their feedback into improving existing features or developing new ones?
Takeshi: To ensure we don’t interpret users’ challenges, pain points, and needs subjectively, we use a framework called the “Opportunity Solution Tree (OST)” to map everything out and present it visually.
Takeshi: The Opportunity Solution Tree (OST) is structured like a tree diagram, where the business goals are listed at the top. Below that, we list the opportunities—such as needs, pain points, and desires—that we’ve identified from user interviews, along with any barriers or issues we’ve observed. The causes of these opportunities are then mapped beneath, and below those, we list potential solutions for each issue.
By using this method, we can break down large, complex challenges into smaller, manageable pieces and tackle them one by one.
It’s a really interesting way to visualize and organize things!
Takeshi: Our team has gained quite a bit of experience with user interviews, but we found it challenging to share the insights we gained with the team and connect them to solutions. That’s when we came across the OST method. Over the past six months to a year, we’ve started experimenting with it, gradually refining the process and adapting it to our team.
Tim: The OST helps us stay focused on the user. By looking at concrete events, we can build hypotheses and come up with solutions. It’s been extremely helpful.
Japanese Manga: A Rich Array of Language Expressions! The Unique Features That Appear in English
While creating the product, have you noticed any unique or interesting aspects about the format of Japanese manga?
Takeshi: I feel that Japanese manga may have emerged because of the Japanese language itself. The format of Japanese manga is deeply rooted in the Japanese language.
For example, when simply translating the existing page layout into English, a problem arises where it becomes unclear who is speaking. When reading manga in Japanese, if there are two people having a conversation with two speech bubbles, we can usually tell right away which character is saying which line.
One reason for this is the unique use of role-specific in Japanese. Role specific-language refers to words that reflect the speaker’s personality or characteristics. For example, it’s more characteristic for an older person to say “Jya” or a woman to add ‘yo ne” at the end of a sentence. Japanese is a language that can easily express these kinds of nuances. This allows manga to omit information that would otherwise be needed to identify the speaker from the art, so the panel layout and speech bubble design work effectively.
In other words, manga design is optimized for the Japanese language. In languages like English, it’s likely more difficult to deduce the speaker’s identity based on the dialogue alone. This becomes a challenge for Japanese learners of English as well, and we’re thinking about whether any future features could help cover this issue.
That’s such an insightful realization!
Takeshi: The Webtoon format, which has been gaining popularity recently (originating from Korea, with a vertical scrolling format), makes this issue much clearer. Since it’s a format developed in non-Japanese language regions, it has been designed to clearly indicate who is speaking, and I was really impressed by that.
Additionally, this isn’t just true for manga, but also anime. One feature of Japanese works is that they tend to have a lot of dialogue. In contrast, overseas works often rely more on visuals and depictions to convey meaning, but in Japanese manga and anime, there’s often more explanation through words. However, this also provides a wide range of input, which I believe is a big advantage when using manga for learning English.
Tim: And it’s not just the dialogue—reading the Japanese version is often more interesting because of all the ‘onomatopoeia’ (words that convey sounds like ‘buzz’ or ‘clatter’) which are more prevalent than in the translated version.
Kazuhiko: Come to think of it, in novels, there’s a tendency to ‘remove’ onomatopoeia. For example, in a scene where someone is eating, the sound ‘mogumogu’ (the sound of munching) would be expressed with stylized text, but it’s much harder to express that in English.
Takeshi: These linguistic differences have a big impact on the product’s expression, and I find that both challenging and interesting. If we can understand these language and cultural differences, we hope to incorporate them into the product development process.
Designing the Experience to Shift from “Seems Learnable” to “Fun”
At the beginning, you mentioned that Langaku is centered around being a language learning app. What do you think is necessary to improve users’ English skills?
Takeshi: Those who get really hooked on Langaku and improve their English skills no longer think of it as “learning English.” They are reading because it’s genuinely fun, and they want to know what happens next. This is the fascinating aspect of extensive reading practice. What starts as an attempt to use spare time productively ends up being an enjoyable activity, where users end up reading hundreds of thousands of words, and one day, they suddenly realize their English skills have improved.
This is exactly what we aim for. At the start, they might feel that they’re “learning English,” but through Langaku, it becomes pure enjoyment, and eventually, their English skills improve dramatically. Creating a product that makes reading manga fun, while still supporting learning, is key to achieving this.
So, the ideal experience design is to start with the “feeling of learning English” and gradually shift to the experience of just “enjoying” reading. What are the current challenges you face in this process?
Takeshi: One challenge we’ve observed is that some users end up looking up too many words or try reading every sentence perfectly. This leads to them feeling frustrated and quitting. They end up thinking “I don’t understand anything” or “this is too hard.” So, our first priority is to address that issue. For example, we’re considering options such as showing a direct translation in addition to the regular translations, or adding an automatic audio reading feature for the speech bubbles. By showing the direct translation, users may better understand why certain phrases are paraphrased in the way they are. And if users can hear the pronunciation of words, it could help them feel more intuitive and reduce the cognitive load of learning.
On the other hand, we also want to strengthen support for users who are motivated to learn vocabulary, idioms, and phrases. This is one area where Langaku has a unique advantage—it’s an app where users can learn a wide range of vocabulary and expressions. By enhancing this aspect, we can satisfy users’ initial motivation of “learning English,” which will help keep them engaged in the long term.
Thank you for the insights. Lastly, could you share your future plans for Langaku?
Takeshi: Moving forward, we want to expand the reach of our product, especially to beginner English learners or those who have hit a learning plateau and are feeling burnt out. Additionally, there are parts of the learning experience that users may find difficult to visualize, so we’re planning to improve our onboarding process, for example, by introducing a manga that helps users understand the learning methods. We’re excited to try out new ideas and continuously improve the product.
Tim: Langaku is the perfect app for anyone looking to learn English. I truly believe it has the potential to change many lives, and I’m excited to have the opportunity to contribute to its development. Our goal is to design the best experience that will have a positive impact on the lives of many users.
Kazuhiko: On the technical side, we plan to improve the accuracy and coverage of our dictionary features. We also want to deepen our understanding of second language acquisition theory (the study of processes and mechanisms for learning a second language) and incorporate that into the product. Currently, Langaku is strengthening its engineering team, so if anyone is interested, we’d be happy to hear from you!
Special thanks to:
Mantra Inc. Corporate Website*If you are interested in Mantra Inc., please check out our Mantra Recruitment for more information.
