glendonc / hanguru

Hanguru: Korean Language Learning Web Application

Introduction

Hanguru is an innovative web application designed to facilitate Korean language learning. It integrates various technologies, including AI and Google Cloud APIs, to offer a comprehensive and interactive learning experience.

Features

Vocab Page

• Create and edit personalized Korean vocab sets using Google Translation API.

Exercise Page

• Practice with generated questions and receive feedback.

Audio Recording

• Generate text for reading practice and record your own voice for review.

Progress Checker

• Track your daily login consistency.

User Account Management

• Register, login, and manage your profile and settings.

Technologies Used

• Backend: Node.js with Express framework, Passport.js for authentication.
• Frontend: HTML, Chakra UI for responsive design.
• Database: MongoDB and Google Cloud.
• AI/ML & Language Processing: OpenAI and Google Cloud tools for Korean language processing.
• Containerization: Docker for deployment consistency.

Environment Setup

Dependencies

• Ensure Docker and MongoDB are installed.

Environment Variables

Create an .env file in the server directory with the following:

MONGO_URI=mongodb://mongo:27017/hanguru
SESSION_SECRET=<your_session_secret>
GOOGLE_API_KEY=<your_google_api_key>
OPENAI_API_KEY=<your_openai_api_key>
GCS_SERVICE_ACCOUNT=gcs-service-account.json
SPEECH_TO_TEXT_SERVICE_ACCOUNT=text-to-speech-service-account.json
GCS_BUCKET_NAME=hanguru-audio-bucket
PORT=8100

Google Cloud Setup

• Obtain and configure the necessary service account JSON files

Docker Network

• Create a network and connect all servers (GCP SQL Proxy, MongoDB, client, and server)

hanguru's Issues

Contextual Help and Tips

Implement contextual tips for language learners by analyzing their input and performance using KoBERT

Gamification Elements for Engagement

Add gamification elements such as points, badges, and levels to encourage learning

Progress Checker - Login Streak

• Display the highest consecutive login count
• Keep track of total days logged in
• User should be able to see streak by switching through the months

Secure User Authentication with JWT/Session Tokens

Implement JWT or session-based authentication to securely manage user sessions

AI-Generated Sentence Creation using OpenAI Model

Leverage KoBERT for creating sentences for user-inputted words and grammar points, ensuring they are diverse and contextually appropriate

Data Sanitization Before Saving to Database

Sanitize user data before saving to prevent NoSQL injection attacks

Implement Direct Deployment

Setup and build workflow files

Set Up MongoDB Indices for User Collection

Create necessary indices on the User collection for optimized queries

Error Handling for User Registration and Login

Improve error responses for better client-side handling

Implement User Login Endpoint

Create an API endpoint to authenticate users and establish a session or token

Study Groups and Language Partner Matching

Implement features to form study groups or match language partners

User Community and Forum

Develop a community forum where users can interact, ask questions, and share knowledge

CI/CD

• Choose CI/CD Tool
• Figure out how pipeline will integrate into project
• Develop continuous deployment strategy

AI-based Study Material Recommendation

Develop a recommendation engine to suggest study material tailored to the user’s level and progress

Progress Tracking System

Create a system to track and display user progress and performance over time

Implement tests for Selenium

Implement Rate Limiting on Authentication Routes

Protect the login and registration endpoints from brute force attacks using rate limiting

Adaptive Testing Algorithms

Develop algorithms to adapt testing based on user's past performance

Real-time Pronunciation Feedback

Implement real-time feedback on user’s pronunciation using speech recognition technology

Implement User Registration Endpoint

Create an API endpoint to register new users, including password hashing and validation

User Input Validation

Validate user input on both registration and login to ensure data integrity

Implement tests for JMeter

Sentence Generation Logic Implementation

Develop the logic for generating sentences using the stored Korean words and grammar points

Language Exercise Interface Creation

Create interfaces for exercises like translation, fill in the blank, and error detection

Email Confirmation on Registration

Send a confirmation email after user registration with a token to verify the account

(Model)-based Grammar Point Identification

Use KoBERT to identify and suggest grammar points within user-generated sentences

Leaderboards and Study Challenges

Introduce leaderboards and periodic challenges to foster a competitive and collaborative environment

Audio Recording and Storage Capability

Implement the ability for users to record their speech and save these recordings

Integrate() for Advanced Sentence Analysis

Incorporate KoBERT to analyze and generate grammatically correct and contextually relevant Korean sentences

Korean Language Input Functionality - Grammar

Allow users to input Korean words and grammar points into a personal dictionary or database