A deep learning project focused on musical instrument classification using Convolutional Neural Networks (CNNs). This project explores the use of different CNN architectures to classify musical instruments from audio data.
The goal of this project is to build and evaluate multiple deep learning models to accurately classify musical instruments. The dataset underwent preprocessing steps including data augmentation and class imbalance handling to improve model generalization and robustness.
| Model | Accuracy |
|---|---|
| GoogleNet | 98.00% |
| MobileNet | 98.00% |
| ResNet | 94.67% |
| VGG16 | 89.33% |
- ✅ Data Augmentation to enrich the dataset and prevent overfitting
- ⚖️ Class Imbalance Handling for balanced class representation
- 🔧 Fine-tuning of pretrained CNN architectures for improved performance
- 🧱 Implemented models:
- AlexNet
- VGG16
- GoogleNet
- ResNet
- MobileNet
- Dataset details and structure are provided in the Documentation folder.
- Preprocessing includes:
- Normalization
- Augmentation
- Balancing classes
This project is licensed under the MIT License.
Musical Instrument Detection, Deep Learning, Convolutional Neural Networks, GoogleNet, MobileNet, ResNet, VGG16, Data Augmentation, Class Imbalance, Fine-tuning, Audio Classification