International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 10 | Oct 2024
p-ISSN: 2395-0072
www.irjet.net
MUSIC SENTIMENT DETECTION PLATFORM Saurabh Chalise1, Sudip Rana2 1 Department of Electronics and Computer, Thapathali Campus, IOE, TU
2 Assistant Professor, Department of Electronics and Computer, Thapathali Campus, IOE, TU
---------------------------------------------------------------------***--------------------------------------------------------------------learning, this project aims to deepen the understanding of Abstract - The consumption of content has significantly this relationship. The project seeks to simulate the brain's increased in the internet age, with music being one of the response to music by composing music autonomously most profitable and widely studied forms of media. Various based on given inputs, showcasing the application of AI in factors, such as sex, music preference, and individual creating music that reflects specific emotional states. This personality, have been considered in the development of a work attempts to reveal the intricate connection between Music Sentiment Detection Platform (MSDP). Deep Neural music and emotions, offering new insights into how Networks (DNN) have been employed for classification, technology can replicate and understand human enabling near-accurate prediction of music emotions. emotional experiences through music. Russell's Two-Dimensional Emotion Model has been used to identify musical characteristics and categorize songs. After 1.1 Motivation classifiers have been trained using the dataset, a collection of unknown songs has been evaluated to assess the model's Scrutinizing the correlation between music and accuracy. Musical features, including tempo, pitch, rhythm, invoked emotion has been our major motivation and harmony, have been analyzed to classify songs into behind the project. Invoked emotion varies from emotional categories. This platform has been shown to be individual to individual. But regardless of anything, highly beneficial for music streaming services, content tags generated by MSDP can be used to further creators, and therapeutic applications by enabling investigation listening habits of individuals and can personalized music recommendations based on listeners' contribute to better the overall listening experience. emotional states. Furthermore, dynamic mood detection has For a long time, psychologists have been attempting to been supported, assisting users in discovering music that comprehend how music affects and preserves aligns with or alters their current emotional experience. emotion. But due to factors such as age, sex, cultural Through its automatic recognition of emotional nuances, background, musical taste, etcetera proper studies the platform has offered a novel approach to enhancing have been hard to perform. With advanced real-time user interaction with musical content. technology, we hope to compile larger data faster and give proper results. On other hand, Music emotion is Key Words: AI, DNN, MSDP, ML, Model, Arousal and biased from person to person. So, predicting the Valance, Music sentiment, Music emotion of the music is a little ambiguous. This project aims to predict the emotion of music using a machine 1. INTRODUCTION learning model. If the whole process were to be done manually it would take a long time and the probability Music has always played a significant role in human for human error is also high. To accurately compute history, influencing culture and emotions across the data, we need the help of AI to reduce the errors civilizations. Today, with easy access to vast music and increase accuracy. libraries, modern listeners interact with music daily. As the volume of musical content continues to grow, 2. LITERATURE REVIEW accurately categorizing music based on emotional responses has become essential. Music is a powerful Research in music emotion analysis has progressed medium capable of evoking strong emotions, and significantly, yet there remains room for improvement. understanding its impact is critical in various domains, Music evokes not only emotional responses but also including content personalization and emotional wellphysiological effects, prompting the use of artificial being. In this digital age, there is a growing interest in intelligence (AI) to evaluate these emotional movements. understanding how music influences emotions and how Machine learning (ML) approaches typically encompass this knowledge can be applied to develop consumer three main tasks: data preparation, training, and products tailored to individual emotional needs. evaluation. Various ML models have been utilized to identify the emotions associated with music, including Artificial intelligence (AI) and advanced computing have Deep Neural Networks (DNNs), Linear Regression, been increasingly employed to explore this phenomenon. Random Forests, and Support Vector Machines (SVMs). Many studies have established a correlation between Research has compared these models regarding their music and emotional responses. By leveraging machine
© 2024, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 261