Real-time Emotion Recognition in Music Using Machine Learning

Thesis Abstract

**Abstract
** This thesis explores the implementation of real-time emotion recognition in music using machine learning techniques. Music has the unique ability to evoke emotions in listeners, making it an interesting domain for studying emotion recognition. Emotion recognition in music is a challenging task due to the subjective nature of emotions and the complex patterns present in music. Machine learning algorithms offer a promising approach to automatically detect emotions in music, enabling applications such as personalized music recommendations, emotion-aware music generation, and mood-based playlist creation. The first part of the thesis focuses on the background of the study, highlighting the significance of emotion recognition in music and the existing challenges in this field. The problem statement identifies the need for accurate and real-time emotion recognition systems to enhance user experience and engagement with music. The objectives of the study include developing a machine learning model capable of accurately detecting emotions in music and implementing a real-time system for emotion recognition. The literature review in Chapter Two provides a comprehensive overview of existing research on emotion recognition in music, machine learning techniques for feature extraction and classification, and relevant datasets and evaluation metrics. The review synthesizes key findings and identifies gaps in the literature that this thesis aims to address. Chapter Three details the research methodology, including data collection and preprocessing, feature extraction, model selection, training and evaluation procedures, and software implementation. The methodology section outlines the steps taken to build and evaluate the emotion recognition system, ensuring transparency and reproducibility of the results. Chapter Four presents a detailed discussion of the findings, including the performance of the machine learning model in detecting emotions in music, the impact of different feature sets on classification accuracy, and the real-time processing capabilities of the system. The discussion section analyzes the strengths and limitations of the proposed approach and suggests future directions for research in this area. Finally, Chapter Five offers a conclusion and summary of the thesis, highlighting the key contributions, implications, and recommendations for future work. The conclusion emphasizes the significance of real-time emotion recognition in music and its potential applications in enhancing user experiences and interactions with music content. In conclusion, this thesis contributes to the field of emotion recognition in music by demonstrating the feasibility of implementing real-time systems using machine learning techniques. The findings provide valuable insights into the challenges and opportunities in this domain, paving the way for further research and innovation in emotion-aware music technologies.