Overview
Miko is a mixed reality language learning experience designed around one core belief: people remember language best when it is connected to real life. Instead of learning from lists and quizzes alone, users learn from the objects, spaces, and moments around them. Using passthrough AR, natural hand interaction, and conversational AI, Miko transforms the user’s home into a living language classroom.
My role focused on shaping the overall product direction, defining interaction models, designing MR-native workflows, and ensuring the experience feels intuitive, encouraging, and emotionally supportive.
The Problem
Most language learning tools teach words in isolation, far removed from real context. This results in:
-
Low memory retention
-
Difficulty applying language in real situations
-
Anxiety around speaking
-
A disconnection between learning and everyday life
Research shows that people remember more when they learn in context and interact physically with concepts. Mixed reality creates the perfect environment for this.
Design Goal
Create a language-learning companion that:
-
Integrates naturally into daily life
-
Uses real objects and real spaces as teaching tools
-
Reduces speaking anxiety through warm, conversational interactions
-
Feels playful, lightweight, and effortless to use
-
Uses MR only where it adds genuine value
Miko - Your Language Buddy
Product Design for Meta Quest - Meta Horizon Start Developer Competition
Learn a new language through daily life, not textbooks. Miko's immersive AR turns your daily surroundings into a language classroom.
Learn From Daily Life
Users simply point at or hold an item and ask, “What is this in Spanish?” Miko identifies the object, teaches the vocabulary, provides examples, and supports pronunciation. This transforms everyday surroundings into opportunities to learn.
Personal Visual Dictionary
Users can say “Save this to my dictionary,” and Miko captures a photo of the real object. Over time, the dictionary becomes a personalized collection of vocabulary tied to lived experiences — improving memory retention.
Learn From Image Mode
A mode for more advanced learners. Users load an image and Miko guides them through describing scenes, building sentences, and understanding grammar through context.
Adaptive Quizzes
Miko generates personalized quizzes based on what the user has previously learned. These reinforce memory while keeping the experience simple and approachable.
Alphabet Mode
A playful experimental mode for exploring vocabulary from A to Z, inspired by the way children learn their first words.
Solution
Design Process
1. Research & Insights
I explored research around embodied cognition and spatial memory, which shows that people retain information better when learning is tied to physical interaction and environment. This insight became foundational to Miko: instead of asking users to imagine situations, mixed reality allows them to learn directly from the spaces and objects they already inhabit.
2. Interaction Design
The interaction design focused on maximizing what mixed reality does best: hands-free, spatial, and context-aware learning. Unlike traditional apps that rely on taps, buttons, and menus, Miko uses natural hand gestures and voice as the primary inputs. Gestures like palm taps and fist actions allow users to navigate, capture images, and trigger learning flows without breaking immersion or switching mental modes.
This approach reduces cognitive load and removes the feeling of “studying.” Learning happens while users move, point, hold objects, and speak naturally, which is fundamentally different from learning on a phone or laptop. The goal was to make language learning feel like part of daily life rather than a task users have to sit down and complete.
3. MR-Native UX
Designing for passthrough MR required a shift away from flat UI assumptions. Instead of fixed screens, I treated the environment as part of the interface. Panels and prompts were placed spatially, anchored to the user’s view in a way that feels supportive rather than intrusive.
I iterated extensively on contrast, depth, scale, and placement to ensure text remained readable across unpredictable real-world backgrounds. Visual clutter was minimized to avoid overwhelming users, especially during speaking practice. Every UI decision was evaluated through the lens of comfort, legibility, and spatial awareness, something traditional 2D interfaces do not need to consider.
4. Prototyping
I built rapid prototypes in Unity to test key flows such as object-based learning, image-based practice, saving vocabulary to a personal dictionary, and quiz interactions. Testing on-device was essential, as MR interactions behave very differently in real environments compared to the editor. Each iteration helped refine timing, placement, and interaction clarity.
What’s Next
The next phase of Miko focuses on deepening the quality of learning through iteration, rather than expanding surface-level features. A key priority is refining the conversational flow to feel even more natural, adaptive, and human. This includes improving how Miko responds to hesitation, mistakes, and follow-up questions, allowing conversations to better mirror how people naturally learn and correct language in real life.
I plan to continue testing and iterating on interaction timing, feedback pacing, and conversational tone to ensure users feel supported rather than corrected. Future iterations will also explore how Miko can better understand user intent and context, adapting explanations and examples based on prior learning history and real-world usage.
The long-term goal is to make Miko feel less like a tool and more like a natural learning companion that grows with the user through daily use.