We dedicated to develop an AI agent capable of classifying music into ten distinct genres with a high level of confidence. By leveraging machine learning, deep learning, and ensemble learning techniques, we aim to create a robust and effective music classifier.
- Duration: April 2024 to July 2024
- Machine Learning Techniques: Implementing a range of machine learning, deep learning, and ensemble learning techniques to enhance classification accuracy.
- Custom Dataset: Creating our own dataset through web scraping, improving upon the existing GTZAN dataset to ensure comprehensive and diverse data.
- Model Evaluation: Thoroughly evaluating our models and employing strategies to avoid overfitting, ensuring the production of the most effective classifier.
- Deployment: Packaging our solution into a Docker image, deployed via Fly.io, making it accessible online for public use.
You can visit this link to experience the agent: https://infra-monitor-patient-mountain-3013.fly.dev:5000/
Or if you want to run the application locally, check the README.md in the source_code folder
This project was made possible through the efforts and dedication of our team, as well as the resources
provided by the open-source community. We hope this classifier
helps you gain better insights into your playlist and proves useful to music enthusiasts worldwide.
We hope you enjoy using our Music Genre Classification AI agent and find it as insightful and helpful as we intended it to be!