🎧 Urban Sound Classification System An AI-based system that classifies environmental sounds using machine learning. This project processes audio files and predicts categories such as sirens, dog barking, drilling, and street noise using the UrbanSound8K dataset.
📌 Overview Urban environments contain a variety of sounds that can be analyzed for smart applications. This project builds a sound classification model that: Takes an audio file as input Extracts meaningful audio features Uses a trained model to classify the sound
Outputs the predicted category 🚀 Features 🎵 Accepts .wav audio input 🔍 Feature extraction using MFCC and spectral features 🤖 Machine Learning model (Random Forest / CNN) 📊 Accurate classification of 10 sound categories 📈 Easy to extend for real-time applications
🧠 Technologies Used
Python Librosa (Audio Processing) NumPy & Pandas Scikit-learn / TensorFlow / Keras Matplotlib
📂 Dataset UrbanSound8K Dataset 8732 labeled sound clips 10 Classes: Air Conditioner Car Horn Children Playing Dog Bark Drilling Engine Idling Gun Shot Jackhammer Siren Street Music
⚙️ Project Workflow Audio Input Load .wav audio file Preprocessing Noise reduction Normalization Feature Extraction MFCC (Mel Frequency Cepstral Coefficients) Chroma Features Spectral Contrast Model Training Train using Random Forest / CNN Prediction Output predicted sound label
🛠️ Installation
Bash
git clone https://github.com/your-username/urban-sound-classifier.git
cd urban-sound-classifier
pip install -r requirements.txt
Output:
Predicted Sound: Siren Confidence: 92%
📁 Project Structure
urban-sound-classifier/ │── data/ │── models/ │── src/ │ ├── preprocess.py │ ├── feature_extraction.py │ ├── train_model.py │ ├── predict.py │── requirements.txt │── README.md
📊 Model Details Algorithm: Random Forest / Convolutional Neural Network Input: Extracted MFCC features Output: Sound class label Evaluation Metrics: Accuracy Confusion Matrix
📌 Applications Smart City Monitoring Traffic Sound Detection Industrial Noise Analysis Emergency Sound Recognition
🔮 Future Improvements Real-time sound detection Mobile application integration Deep learning improvements (LSTM, CNN tuning) Web deployment using Flask / Streamlit
👨💻 Author Mohamed Ansari
⭐ Contributing Contributions are welcome! Feel free to fork the repository and submit a pull request.
📜 License This project is licensed under the MIT License.