🎙️ Speech Emotion Recognition (FastAPI + Deep Learning)

This project is a Speech Emotion Recognition (SER) system that predicts human emotions from audio files (.wav) using a trained deep learning model.
The backend is powered by FastAPI, and users can upload speech samples to get real-time emotion predictions along with confidence scores.

---

Demo Pictures

---

🚀 Features

• Upload a .wav audio file via API or frontend.
• Automatic preprocessing (mel spectrogram generation).
• Emotion prediction using a trained Keras model (model.keras).
• Returns top emotion with a confidence score.
• REST API built with FastAPI (interactive Swagger UI at /docs).
• Optional frontend integration for easy testing.

---

🧠 Model Details

• Input shape: (128, 345, 1)
• Audio format: .wav (mono, 4 seconds, 44100 Hz sample rate)
• Preprocessing:
  • Converts audio to mel spectrogram
  • Pads shorter clips with zeros
  • Normalization & standardization applied
• Output layer: Softmax probabilities
• Classes: ['anger', 'calm', 'disgust', 'fear', 'happiness', 'neutral', 'sadness', 'surprise']

⚙️ Installation & Setup

1. Clone the repository

git clone https://github.com/yourusername/speech-emotion-recognition.git
cd speech-emotion-recognition

---

2. Create a virtual environment

source venv/bin/activate        # Linux / Mac
venv\Scripts\activate           # Windows

---

3. Install dependencies

pip install -r requirements.txt

---

4. Run the FastAPI server

uvicorn app:app --reload

5. Open in browser

API Docs: http://127.0.0.1:8000/docs

Web App (if enabled): http://127.0.0.1:8000

---

API Usage

Endpoint: /predict

Method: POST Description: Upload a .wav file to predict emotion.

Example (cURL) curl -X 'POST'
'http://127.0.0.1:8000/predict'
-F 'file=@sample.wav'

Response Example:
{
  "predicted_emotion": "happiness",
  "confidence": 0.92,
  "class_probabilities": {
    "anger": 0.01,
    "calm": 0.02,
    "disgust": 0.03,
    "fear": 0.05,
    "happiness": 0.92,
    "neutral": 0.01,
    "sadness": 0.01,
    "surprise": 0.03
  }

Technologies Used

• Python 3.10+

• TensorFlow / Keras

• FastAPI

• Librosa

• NumPy

• Uvicorn

• Jinja2 (for frontend)

📊 Future Improvements

Add support for real-time microphone recording.

Improve noise robustness.

Deploy on cloud (Render / Hugging Face Spaces).

Add waveform and spectrogram visualizations.

👨‍💻 Author

Ali Ahmad

AI/ML Engineer | Deep Learning Enthusiast

📍 The Islamia University of Bahawalpur