🚀 VirtueSense: AI-Powered Interview Practice Platform

VirtueSense is an intelligent interview simulation platform that analyzes facial expressions, voice tone, and speech clarity to help users improve communication skills and confidence in interviews.
It combines Flask, Node.js, React + Vite, OpenAI, and DeepFace — all containerized using Docker for easy setup and deployment.

---

🧠 Features

• 🎤 Real-time speech-to-text analysis using SpeechRecognition and OpenAI
• 😊 Emotion & facial expression detection with DeepFace and OpenCV
• 🧩 Unified Flask + Node.js + React microservice architecture via Docker
• 🔊 Supports audio-based and text-based feedback
• 📊 Provides structured insights into clarity, tone, and confidence
• 🐳 Dockerized for instant environment setup with one command

---

📦 Prerequisites

Before you begin, make sure you have the following installed:

Dependency	Minimum Version	Purpose
Docker	Latest	Containerization of all services
Git	—	Clone repository

---

🛠️ Step 1: Clone Repository

git clone https://github.com/PaperKidz/IntervueSense.git
cd IntervueSense

---

⚙️ Step 2: Environment Configuration

Create .env files for each service using the provided examples:

# From project root
cp Backend/.env.example Backend/.env
cp backend-nodejs/.env.example backend-nodejs/.env
cp Frontend/.env.example Frontend/.env

Then open each .env file and fill in your credentials.
For example, in Backend/.env:

OPENAI_API_KEY=sk-your-secret-key

Make sure all .env files contain correct paths, ports, and API keys.

---

🐳 Step 3: Build & Run with Docker

Run the following command from the project root:

docker-compose up --build

Docker will automatically:

1. Build images for Flask, Node.js, and React (Vite)
2. Set up network and environment variables
3. Start Nginx as a reverse proxy routing between frontend and backend containers

Once the build completes, open your browser:

http://localhost

---

⚡ Project Structure

VirtueSense/
│
├── Backend/                # Flask backend (Python)
│   ├── app.py
│   ├── requirements.txt
│   ├── .env.example
│
├── backend-nodejs/         # Node.js microservice backend
│   ├── server.js
│   ├── package.json
│   ├── .env.example
│
├── Frontend/               # React + Vite frontend
│   ├── vite.config.js
│   ├── package.json
│   ├── .env.example
│
├── nginx/
│   └── nginx.conf          # Reverse proxy configuration
│
├── docker-compose.yml
└── README.md

---

🌍 Default Service URLs

Service	URL	Description
Frontend (React)	http://localhost	Main user interface
Flask Backend	http://localhost:5000	AI & DeepFace analysis
Node.js Backend	http://localhost:4000	Evaluation, progress APIs
Nginx	http://localhost	Reverse proxy to all services

---

🔌 API Endpoints (Flask Backend)

Endpoint	Method	Description
/api/analyze_face	POST	Analyze facial emotion from base64 or uploaded image
/api/analyze_audio	POST	Process audio input for tone & emotion
/api/transcribe	POST	Convert user audio speech to text
/api/feedback	POST	Generate AI-based feedback

Example request:

curl -X POST http://localhost/api/analyze_face \
-H "Content-Type: application/json" \
-d '{"image": "data:image/jpeg;base64,..."}'

---

🧠 Tech Stack

• Frontend: React + Vite, TailwindCSS
• Backend (AI): Flask, DeepFace, OpenCV, OpenAI API
• Backend (Logic): Node.js + Express
• Speech Processing: SpeechRecognition, Pydub, FFmpeg
• Reverse Proxy: Nginx
• Containerization: Docker & Docker Compose

---

🧰 Useful Commands

Command	Description
docker-compose up --build	Build and start all containers
docker-compose down	Stop and remove all containers
docker-compose logs -f	View live logs from all services
docker system prune -a	Clean up all unused Docker data

---

🧩 Troubleshooting

• Ports already in use?
  Stop existing services using those ports or modify them in .env and docker-compose.yml.

• Changes not reflecting?
  Run docker-compose up --build again to rebuild with latest updates.

• Nginx 502 Bad Gateway?
  Ensure both backends are healthy (docker ps) and reachable inside the network.

---

🏁 Done!

Visit http://localhost to start your AI-powered interview simulation.
Docker will take care of running everything — no manual setup needed.