A Study on an Effective Music Distribution System for Online Digital Music Industry

Abstract

Due to the vast improvements in the Internet and telecommunication technologies and digital multimedia platforms, the music industry has already been moving into online space successfully these days. Services of digital music market such as online streaming, downloading, and online music stores are extremely popular among young consumers. Online music distribution brings several benefits such as less coordination and distribution costs for distributors, and cost and time effectiveness for buyers. Along with the benefits, it has also brought some challenges: how to securely distribute songs online, how to control illegal access to songs, how to share teaser/sample music for potential buyers to taste the songs. The proposed system in this thesis is intended to satisfy the above requirements of online digital music industry. The proposed system is developed based on the interesting relation between the Discrete Wavelet Transform (DWT) and the human auditory system. To understand the proposed method properly, detail and approximation coefficients resulting from the DWT decomposition of an audio/music signal should be clearly understood. The details are the high-frequency components of an audio signal, while the approximations correspond to the low-frequency components. The normal human ear is usually most sensitive in the lowfrequency range. Thus, the approximation coefficients are more important for audio quality. Slight modification on those coefficients will severely degrade the audio quality, and whereas the details are not that much sensitive to the human ear. Based on the above nature of the DWT, the proposed system is developed as follows. An audio/music signal is first broken down into different DWT wavelet layers. Then, the proposed audio scrambling method is applied on each layer with different keys from a pre-generated key table. The resulting music signal with scrambled detail coefficients has degraded quality but is still understandable the lyrics and the essence of the music. Thus, it can be used as teaser for potential buyers. Otherwise, if stronger security is needed for music distribution to buyers, only the approximations (or) all coefficients can be scrambled. This will yield the music with very low quality which is nearly noisy. Without knowing the keys, anyone can never restore the original music quality. In this way, unauthorized access to music can also be efficiently controlled.iv The proposed system is implemented in MATLAB R2017a. Experimental results show that the proposed method is simple to implement and very effective in terms of low computational complexity and fast execution time. In addition, it supports flexible direct control on the music quality. Thus, the proposed system is applicable for online digital music industry.