Journal of Engineering and Applied Sciences

Year: 2018
Volume: 13
Issue: 8 SI
Page No. 6378 - 6389

Highly Efficient and Robust Audio Identification and Analytics System to Secure Royalty Payments for Song Artists

Authors : Tharika Madurapperuma, Gothami Abayawickrama, Nesara Dissanayake, V.B. Wijesuriya and K.l. Jayaratne

Abstract: Radio is yet one of the most popular broadcast media in many emerging regions and operates as one of the principle sources of income for licensors and publishers, collectively termed as owners of copyrighted music. Royalty payments are legal obligations towards owners of copyrighted music under intellectual property rights legislations. Independent conjoint monitoring of copyrighted music being broadcast over every radio station on each day is an intricate requisite for upholding the said legislations of a country or region and therefore, requires automated techniques that are efficient, scalable and robust to both radio and content noise. We consider the development of an automated radio broadcast audio monitoring system with identification and analytics capabilities to assist collection of royalty payments from radio music licensees for copyrighted music, specifically for songs. First, we pre-process a broadcast radio stream to identify song segments, aka objects via. an efficient onset detection mechanism. Next, we perform efficient hashing of identified objects in the frequency domain and compare generated hashes with those in the entries of a precompiled copyrighted song database using an efficient hash matching technique. Subsequently, our system stores broadcast data of the identified music objects (e.g., timestamp, name of lyricist, etc.) in the database for later use from an analytic application.

How to cite this article:

Tharika Madurapperuma, Gothami Abayawickrama, Nesara Dissanayake, V.B. Wijesuriya and K.l. Jayaratne, 2018. Highly Efficient and Robust Audio Identification and Analytics System to Secure Royalty Payments for Song Artists. Journal of Engineering and Applied Sciences, 13: 6378-6389.

Design and power by Medwell Web Development Team. © Medwell Publishing 2023 All Rights Reserved