AI English Study Partner

English | 繁體中文

A multimodal AI-powered English learning chatbot built on the LINE platform, designed to help users engage in contextual English conversation with seamless interaction via text, speech, and image.

Table of contents

• Key Features

• How To Use

• Project Structure

• License

Key Features

• ✅ Multimodal Chatbot: Combines text, voice, and image input/output to create immersive learning scenarios.
• 🧠 Integrated Deep Learning Models:
  • LLM (Large Language Model): Handles conversation generation.
  • STT (Speech-to-Text): Converts user speech into text.
  • TTS (Text-to-Speech): Reads AI responses aloud.
  • VLM (Vision-Language Model): Processes image-based questions or inputs.
• 📚 RAG (Retrieval-Augmented Generation): Enhances accuracy and informativeness of generated responses using vector database.
• 🔄 Scheduled Push Notifications: Keeps learners engaged and builds daily habits.
• 💬 LINE Bot Interface: No need to download apps—just chat on LINE.
• ⚙️ One-click Environment Setup: Run install.py to install all required dependencies.

How To Use

🌐 1. Set up ngrok (Expose your local server to the internet)

LINE Bot webhook requires a public HTTPS URL. Ngrok helps you create a secure tunnel from your local machine:

Note the HTTPS URL generated you will use this as your webhook URL.

🔔 2. Register and configure your LINE Bot

Go to the LINE Developers Console, create a Provider and then a Messaging API Channel.

Find and copy your LINE_CHANNEL_ACCESS_TOKEN and LINE_CHANNEL_SECRET.

Enable Use webhook.

Set your webhook URL to your ngrok HTTPS URL.

Add your LINE Bot as a friend via the QR code in the LINE Developers Console.

🐍 3. Prepare Python environment and install dependencies

Python version: 3.12 (Recommanded)

Initial setup time: ~10 minutes (via install.py)

git clone https://github.com/your-username/AIEnglishStudyPartner.git
cd AIEnglishStudyPartner

python -m venv venv
# Activate virtual environment:
# On macOS/Linux:
source venv/bin/activate
# On Windows:
venv\Scripts\activate

python install.py

The install script will:

• Install base dependencies

• Install CUDA-enabled PyTorch

• Install MeloTTS from GitHub

• Download UniDic Japanese dictionary

🔑 4. Configure your environment variables

Copy .env.example and fill in your tokens:

cp .env.example .env

Make sure the following variables are set correctly in .env:

• NGROK_TOKEN

• LINE_CHANNEL_ACCESS_TOKEN

• LINE_CHANNEL_SECRET

• HF_TOKEN

Project Structure

AIEnglishStudyPartner/
├── install.py               # One-click install script
├── .env.example             # Environment variable template
├── main.py                  # Entry point (LINE bot server)
├── requirements.txt         # Python dependencies
├── models/                  # LLM, STT, TTS, VLM integration
│   ├── llm.py
│   ├── stt.py
│   ├── tts.py
│   ├── vlm.py
│   ├── rag.py
│   ├── relationalDB.py
│   ├── schemas.py
│   └── __init__.py
├── db
│   ├── vector_db/           # Vector store for RAG
│   ├── relational_db.db
├── static/                  # Static files (images, audio)
└── templates/               # HTML templates 

License

Apache-2.0 license

(Back to top)