Share on facebook
Share on twitter
Share on linkedin
Share on skype
Share on email
Share on reddit
Share on facebook
Share on twitter
Share on linkedin
Share on skype
Share on email
Share on reddit

Analysis Genre of Bangla Music: A Machine Classification Learning Approach

 

Authors : Abhijit Bhowmik, AZM EhteshamChowdhury
Journal Info : — AIUB Journal of Science and Engineering.vol 18,Issue 2, 2019

 

Abstract : The necessity for designing autonomous indexing tools to establish expressive and efficient means of describing musical media content is well recognized. Music genre classification systems are significant to manage and use music databases. This research paper proposes an enhanced method to automatically classify music into different genre using a machine learning approach and presents the insight and results of the application of the proposed scheme to the classification of a large set of The Bangla music content, a South-East Asian language rich with a variety of music genres developed over many centuries. Building upon musical feature extraction and decision-making techniques, we propose new features and procedures to achieve enhanced accuracy. We demonstrate the efficacy of the proposed method by extracting features from a dataset of hundreds of The Bangla music pieces and testing the automatic classification decisions. This is the first development of an automated classification technique applied specifically to the Bangla music to the best of our knowledge, while the superior accuracy of the method makes it universally applicable.

Journal link : Click Here 

Most Recent Publications