INSTINCTIVE MUSIC GENRE DETECTION AND CATEGORIZATION OF AUDIO DATA USING MACHINE LEARNING

Abstract

People's mental health is greatly influenced by music. Music serves as a bond that links communities together by bringing people with common interests together. As a result of people's changing mindsets, music has now become a business. A person may go a day without interacting with others or wishing for a friend nowadays, but they cannot go a day without listening to music. Songs are divided into a variety of genres. Using machine learning algorithms, we can deliver a categorized list of music to the smartphone user. The structure of music, as well as how humans perceive and understand it, is discussed in connection to musical style categorization. For this experiment, we focused on combining data from audio signals rather than data from several sources. The entire project includes a comprehensive machine learning technique for automatically categorizing musical genres based on audio inputs. This ensemble approach suggests that combining different types of domain-based audio characteristics might improve classification accuracy. We'll use machine learning techniques to build models that can categorise audio recordings into different genres. After our trained model has been trained, we shall evaluate its performance. The goal of the research is to develop a machine learning system that outperforms existing music genre predicting methods. In this research, we used the GTZAN dataset to train a number of classification models. We compared and documented the results of all of these models in terms of predicted accuracy.