Bird Species Classifier Web Application 🦜

A modern web application for real-time bird species classification from audio recordings using deep learning models.

Features ✨

• Upload and process audio files (supports .wav, .mp3, and .flac formats)
• Multiple pre-trained models to choose from:
  • MobileNetV2-based models (Chroma and CQT variants)
  • VGG16-based models (Chroma and CQT variants)
• Real-time audio feature visualization:
  • Mel Spectrogram
  • Constant-Q Transform (CQT)
  • Mel-frequency Cepstral Coefficients (MFCC)
  • Chromagram
• Window-based prediction system for temporal analysis
• Responsive and intuitive user interface
• Efficient processing with background noise handling

Supported Bird Species 🐦

The system can classify 21 different categories including:

• Acrocephalus arundinaceus
• Acrocephalus melanopogon
• Acrocephalus scirpaceus
• Alcedo atthis
• Anas platyrhynchos
• And many more...

Installation 🚀

1. Clone the repository:

git clone https://github.com/yourusername/bird-species-classifier.git
cd bird-species-classifier

2. Create and activate a virtual environment (recommended):

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install the required packages:

pip install flask numpy pandas librosa matplotlib seaborn tensorflow scikit-learn flask-session

4. Create the necessary directories:

mkdir uploads
mkdir static/css
mkdir static/js
mkdir templates
mkdir models

5. Place your trained models in the models directory with the following naming convention:

• ChromaMNV21.keras
• ChromaMNV22.keras
• CQTMNV23.keras
• CQTMNV24.keras
• ChromaVGG1.keras
• ChromaVGG2.keras
• CQTVGG3.keras
• CQTVGG4.keras

Usage 💻

1. Start the Flask application:

python app.py

2. Open your web browser and navigate to http://localhost:5000

3. Upload an audio file and select a model for classification

4. View the predictions and analyze the audio features for each time window

Technical Details 🔧

Backend

• Flask web framework
• TensorFlow for model inference
• Librosa for audio processing
• NumPy and Pandas for data handling
• Matplotlib for feature visualization

Frontend

• Pure HTML, CSS, and JavaScript
• Responsive design with modern UI components
• Real-time feature visualization
• Interactive window selection

File Structure 📁

bird-species-classifier/
├── app.py                 # Flask application
├── static/
│   ├── css/
│   │   └── test.css      # Styles
│   └── js/
│       └── trial.js      # Frontend logic
├── templates/
│   └── index.html        # Main page template
├── models/               # Trained models
└── uploads/             # Temporary file storage

Contributing 🤝

Contributions are welcome! Please feel free to submit a Pull Request.

License 📄

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments 🙏

• Thanks to all contributors and testers
• Bird sound datasets providers
• Deep learning model architecture developers