Speak-English-AI-Agent

🏗️ Node.js Backend for AI-Powered Language Learning

This repository contains the backend for an AI-powered English-speaking. The backend is built with Node.js and integrates an open-source LLM (Large Language Model) for real-time speech evaluation, pronunciation correction, and fluency feedback.

🚀 Tech Stack

Node.js - Backend runtime
Express.js - Server framework
Ollama (LLM API) - AI model for speech and text analysis
Docker - Containerized deployment

📦 Installation & Setup

Prerequisites

Ensure you have the following installed:

Node.js (v16+ recommended) → Download Node.js
Docker → Download Docker
Ollama (LLM API) running on localhost:11434

1️⃣ Clone the Repository

git clone https://github.com/vinodnextcoder/Speak-english-ai-agent.git
cd Speak-english-ai-agent

2️⃣ Install Dependencies

npm install

3️⃣ Set Up Environment Variables

Create a .env file in the root directory and add:

PORT=5000
OLLAMA_API_URL=http://localhost:11434/api/generate

4️⃣ Start the Server

npm start

The server should now be running at http://localhost:5000.

📌 Running Mistral 7B with Ollama in Docker

This repository provides an easy-to-use Docker setup for running Mistral 7B using Ollama.

✅ Features

✔️ Run Mistral 7B locally with a single command.
✔️ Pull Ollama and the model directly from the registry.
✔️ Lightweight & efficient setup.

1️⃣ Install Docker

If you haven't installed Docker, get it here:
🔗 Download Docker

Verify installation:

docker --version

2️⃣ Pull the Ollama Docker Image

docker pull ollama/ollama

3️⃣ Run the Ollama Container

docker run -d --name ollama-container -p 11434:11434 ollama/ollama

-d → Run in the background.
--name ollama-container → Assigns a name.
-p 11434:11434 → Exposes Ollama API.

Check if the container is running:

docker ps

4️⃣ Pull the Mistral 7B Model

docker exec -it ollama-container ollama pull mistral

📡 API Endpoints

1️⃣ Analyze Speech/Text

Endpoint: POST /api/analyze
Description: Processes user speech or text and returns AI feedback.
Request Body:
```
{
  "text": "I goes to school."
}
```

Response:

{
  "correctedText": "I go to school.",
  "feedback": "Verb conjugation corrected."
}

🤝 Contributing

We welcome contributions! Feel free to open issues and submit pull requests.

📜 License

MIT License

This backend is designed to be lightweight, efficient, and scalable for real-time AI-powered language learning. 🚀