MUSIC STREAMING APP WITH DEEP LEARNING TECHNIQUES

Abstract

This study presents the development of a music classification and recommendation system for web applications, employing deep learning and visualization methodologies to address the demand for refined audio recommendation systems. Through research and experimentation, high accuracy in music genre classification was achieved. Building a CNN model and employing t-SNE visualization resulted in a clear clustering of audio files. Coordinated of individual songs were used in the recommendation system logic.