🚀 Real-time Emotion Detection System using Face + Voice AI with an intelligent assistant, analytics dashboard, and mood-based responses.
Emotion-AI is a multi-modal AI system that detects human emotions using:
- 📷 Facial expressions (Computer Vision)
- 🎤 Voice signals (Audio Processing)
It combines both to generate:
- 🤖 Smart assistant responses
- 🎵 Mood-based music suggestions
- 📊 Real-time emotion analytics
-
🎭 Face Emotion Detection
- Deep learning model using DeepFace
- Real-time webcam processing
-
🎙️ Voice Emotion Recognition
- Audio feature extraction (MFCC, etc.)
- ML-based classification
-
🤖 AI Assistant
- Emotion-aware responses
- Text-to-Speech output
-
🎵 Music Recommendation
- Plays songs based on detected mood
-
📊 Analytics Dashboard
- Emotion trends
- Visualization of mood patterns
-
⚡ Real-time Processing
- Live face + voice detection
| Category | Technologies Used |
|---|---|
| Programming | Python |
| Computer Vision | OpenCV, DeepFace |
| Machine Learning | Scikit-learn |
| Deep Learning | TensorFlow / Keras |
| Audio Processing | Librosa, SoundDevice |
| Speech Recognition | SpeechRecognition |
| Text-to-Speech | pyttsx3 |
| Visualization | Matplotlib, Seaborn |
| UI / Dashboard | Streamlit |
| Assistant API | OpenAI (optional) |
Emotion-AI/ │ ├── modules/ # Core logic (AI + assistant + audio) ├── training/ # Model training scripts ├── utils/ # Helper functions & analytics │ ├── main.py # Main entry point ├── run_dashboard.py # Streamlit dashboard ├── requirements.txt # Dependencies ├── .gitignore # Ignored files
git clone https://github.com/krishbaloria11/Emotion-AI.git
cd Emotion-AIpip install -r requirements.txtpython main.py
streamlit run run_dashboard.py 📊 Dataset
FER2013 → Face Emotion Dataset RAVDESS → Voice Emotion Dataset
🎯 Model Performance Model Type Accuracy Face Model ~65–70% Voice Model ~80–85%
❌ No user data is stored ❌ No images or videos are saved 🎤 Audio is processed in real-time 📷 Webcam feed is not recorded 🌐 Voice recognition may use external APIs (Google Speech API)
Ensure microphone and webcam permissions are enabled Some features may require internet (speech recognition) OpenAI integration is optional and disabled by default
🌐 Web App Deployment (Streamlit / Flask) 📱 Mobile Application 🎯 Improved accuracy (90%+) ☁️ Cloud-based inference 🧠 Emotion history tracking
Contributions are welcome! Feel free to fork the repo and submit a pull request.
If you like this project:
👉 Give it a STAR ⭐ on GitHub 👉 Share it with others
Krish Baloria GitHub: https://github.com/krishbaloria11
Jatin Github: https://github.com/Jatin-Verma11
This project is for educational purposes.