Title :
Using beat notation for enhancement of chord sheet music document similarity
Author :
Chaisup Wongsaroj;Nakornthip Prompoon;Athasit Surarerks
Author_Institution :
Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand
Abstract :
Most sheet music websites provide chord sheet music document that has only simple text. The system should provide music information that is easy to understand and allows users to search and view a document in many aspects. However, music recommendation and searching capability based on music similarity of chord sequences in documents are challenging features for the system. In this paper, we present a notation of beat information to specify chord length in chord sheet music and its approach for the similarity calculation in three levels: chord, sequence, and music. This approach focuses on beat information of chord sequences in a song. Then, music similarity value between two songs can be determined by using the aggregation of chord sequence similarity values. This work emphasizes the significance of the beat information in chord sequences that can improve the effectiveness of similarity measure. By the experimental results, our music similarity measure was improved about 32% higher in precision after beat information was included.
Keywords :
"Music","Information retrieval","Data mining","Warehousing","Runtime","Harmonic analysis","Frequency modulation"
Conference_Titel :
Software Engineering and Service Science (ICSESS), 2015 6th IEEE International Conference on
Print_ISBN :
978-1-4799-8352-0
Electronic_ISBN :
2327-0594
DOI :
10.1109/ICSESS.2015.7339202