JLPT Micro Mobile Learning APP

A mobile microlearning application for Japanese learners, built with Flutter and Supabase.

JLPT Micro combines vocabulary, grammar, reading, spaced repetition, and learning progress tracking into a lightweight daily learning experience. The project was developed as a mobile learning course project and explores how microlearning and spaced repetition can be integrated into a cross-platform application.

---

App Screenshots

Actual application screenshots showing the four main learning areas in dark mode, with a Traditional Chinese interface and Japanese learning content.

Daily Vocabulary	Grammar Learning
Review Japanese vocabulary with flashcards and track daily task progress.	Browse core grammar patterns and their Chinese meanings.

Reading & Listening	Learning Dashboard
Read short Japanese articles with narration and translation controls.	View the JLPT target, learning streak, weekly activity, and learned items.

---

Project Overview

Traditional language-learning applications often contain large amounts of learning material, which can make it difficult for learners to maintain a consistent study routine.

JLPT Micro is designed around the concept of short and repeatable daily learning sessions. Users can select a target JLPT level and complete daily vocabulary, grammar, and reading tasks while the application records their progress in the cloud.

The system also applies a simplified SM-2 spaced repetition algorithm to vocabulary review. Based on the learner's response, the system dynamically determines when a word should appear again.

---

Features

Daily Vocabulary Learning

• Swipe-based flashcard interface
• Separate new words and review words
• Tap cards to view example sentences and translations
• Swipe right for remembered words and left for unfamiliar words
• Automatic review scheduling using an SM-2 based spaced repetition algorithm
• Vocabulary familiarity and review interval tracking
• Bookmark important vocabulary
• Continue learning with additional cards after completing the daily task

Grammar Learning

• Daily grammar learning and review
• Grammar explanations and example sentences
• Learning progress tracking
• Japanese pronunciation using Text-to-Speech (TTS)
• Separation between new grammar points and review items

Reading & Listening Practice

• Japanese news-style reading materials
• Japanese Text-to-Speech playback
• Chinese translation and key vocabulary support
• Add useful vocabulary from reading materials into the vocabulary learning system
• Track completed reading materials

Learning Dashboard

• JLPT target level configuration
• Consecutive learning-day streak
• Mastered vocabulary count
• Mastered grammar count
• Recent 7-day learning statistics
• Visualized learning data using charts

Additional Learning Features

• Vocabulary bookmarks
• Mastered vocabulary collection
• Mastered grammar collection
• Quiz system
• Achievement / badge system
• Daily learning records

User System

• Email and password registration
• User login / logout
• Password reset
• Persistent login session
• Cloud synchronization of learning progress
• Per-user data isolation with Row Level Security (RLS)

User Experience

• First-time onboarding flow
• Loading skeleton animations
• Light / Dark mode following system settings
• Responsive Flutter Material UI
• Local settings cache for improved usability

---

Spaced Repetition

Vocabulary learning is managed using an SM-2 inspired spaced repetition mechanism.

Each vocabulary item stores information such as:

• Familiarity
• Ease factor
• Review interval
• Next review date
• Number of appearances
• Number of correct responses

When a learner remembers a word, its review interval gradually increases.

When a learner marks a word as unfamiliar, its familiarity level is reset and the word is scheduled to appear again sooner.

This allows JLPT Micro to provide a more personalized learning sequence instead of displaying vocabulary in a fixed order.

---

System Architecture

┌──────────────────────────────┐
│        Flutter Client        │
│        Dart / Material UI    │
└──────────────┬───────────────┘
               │
               │ supabase_flutter
               ▼
┌──────────────────────────────┐
│           Supabase           │
│                              │
│  Authentication             │
│  Data API                   │
│  Row Level Security (RLS)   │
└──────────────┬───────────────┘
               │
               ▼
┌──────────────────────────────┐
│      PostgreSQL Database     │
│                              │
│  Vocabulary / Grammar       │
│  Learning Progress          │
│  Reading Progress           │
│  Daily Sessions             │
│  User Profiles              │
└──────────────────────────────┘

The application does not require a separately maintained application server. Flutter communicates with Supabase through the official Flutter client, while authentication, database access, and access control are handled through Supabase.

---

Tech Stack

Category	Technology
Frontend	Flutter
Programming Language	Dart
UI Framework	Flutter Material
Backend Service	Supabase
Authentication	Supabase Auth
Database	PostgreSQL
Database Language	SQL
Security	PostgreSQL Row Level Security (RLS)
Cloud Client	supabase_flutter
Local Storage	shared_preferences
Text-to-Speech	flutter_tts
Data Visualization	fl_chart
Flashcard Interaction	flutter_card_swiper
Loading UI	shimmer
Japanese Font	google_fonts

---

Database Design

The Supabase database contains the following main tables:

Table	Purpose
vocabulary_bank	Vocabulary content
user_word_progress	User vocabulary learning and SRS progress
grammars	Grammar learning content
user_grammar_progress	User grammar learning progress
news	Reading materials
news_vocab	Important vocabulary associated with reading materials
user_news_progress	User reading progress
daily_sessions	Daily learning activity records
user_profiles	User settings and target JLPT level

User-specific records are protected using Row Level Security, ensuring that authenticated users can only access their own learning data.

---

Project Structure

jlpt_micro/
├── lib/
│   ├── main.dart
│   ├── models/              # Application data models
│   ├── screens/             # UI pages
│   │   ├── auth/
│   │   ├── onboarding/
│   │   ├── splash/
│   │   ├── home/
│   │   ├── grammar/
│   │   ├── reading/
│   │   ├── profile/
│   │   ├── mastered/
│   │   ├── quiz/
│   │   └── achievements/
│   ├── services/            # Supabase and learning logic
│   └── utils/               # Theme and reusable utilities
│
├── android/
├── ios/
├── web/
├── windows/
├── macos/
├── linux/
│
├── supabase_setup.sql
├── supabase_advanced.sql
├── supabase_fixes.sql
├── supabase_seed_batch2.sql
├── supabase_seed_batch3.sql
├── REQUIREMENTS.md
└── pubspec.yaml

The application separates UI, data models, and service logic, making the project easier to maintain and extend.

---

Getting Started

Requirements

• Flutter SDK >= 3.41.x
• Dart SDK >= 3.11.4
• JDK 17+
• Supabase account

1. Clone the repository

git clone https://github.com/BlankTsai/jlpt_micro.git
cd jlpt_micro

2. Install dependencies

flutter pub get

3. Configure Supabase

Create the local secrets file:

cp lib/secrets.example.dart lib/secrets.dart

Then configure:

const supabaseUrl = 'YOUR_SUPABASE_URL';
const supabaseAnonKey = 'YOUR_SUPABASE_ANON_KEY';

4. Initialize the database

Run the SQL scripts in the Supabase SQL Editor according to the setup instructions in REQUIREMENTS.md.

5. Run the application

flutter run

To build an Android APK:

flutter build apk --release

---

Security

• Supabase credentials are separated from the source code through secrets.dart
• Secret configuration files are excluded from Git
• User authentication is handled by Supabase Auth
• User learning records are protected using Row Level Security
• User passwords are never stored directly by the Flutter client

---

What I Learned

Through this project, I practiced building a complete mobile application from UI design to cloud data persistence, including:

• Cross-platform application development with Flutter
• Dart asynchronous programming
• Relational database design
• SQL and data modeling
• User authentication
• Cloud database integration
• Row-level access control
• Separation of UI and service logic
• User learning-progress modeling
• Implementation of a spaced repetition algorithm
• Learning-data visualization
• Mobile interaction and user-experience design

More importantly, the project allowed me to combine software engineering with digital learning design, transforming the concept of microlearning into an actual working system.

---

Current Development Status

The core learning workflow, authentication, learning progress tracking, spaced repetition, reading practice, statistics, quiz, and achievement systems have been implemented.

The application supports selecting JLPT levels from N5 to N1, while the current bundled learning dataset primarily focuses on selected levels such as N5 and N3.

Future extensions may include:

• Expanding the JLPT vocabulary and grammar dataset
• Improving the spaced-repetition model
• Adding more detailed learning analytics
• Expanding automated testing
• Improving cross-platform UI adaptation
• Introducing personalized learning recommendations

---

Author

BlankTsai

National Taiwan Normal University

GitHub: BlankTsai