International Journal of Computer Applications (0975 – 8887) Volume 70– No.26, May 2013 5 Analysis of Music Signal Compression with Compressive Sensing Urvashi P Shukla PG Student, EC SCET, Surat INDIA Niteen B Patel Professor, EC SCET, Surat INDIA Amit M Joshi Research Scholar, EC SVNIT, Surat INDIA ABSTRACT Music signal processing has matured over the years. Most of the techniques applied to music signal were originally developed for speech signal; the results observed have been quite satisfactory. In music domain, there are lot many challenges faced with respect to the storage and transmission of large data base. In this paper, we try to address the above stated problems with help of growing compressive sensing field. The aim is to extract the features of the signal with aid of the basis which will in turn reduce the number of samples to be stored or transmitted. Also, our goal is to successfully demonstrate the recovery of the signal without hampering the inner characteristics and the melody of the signal. General Terms Compressive sensing, Algorithm, Simulation Keywords Basis matrices, Discrete Fourier Transform, Measurement matrix, Music signal, Recovery algorithm. 1. INTRODUCTION Music is a ubiquitous and vital part of the many lives worldwide. Musical creations and performances are among the most complex and intricate of our cultural artifacts, and the emotional power of music can touch us in surprising and profound ways [1].Music has quiet large span range of forms and styles like simple, unaccompanied folk songs, orchestras and other large ensembles, to a minutely constructed piece of electronic music resulting from months of work in the studio. With a great amount of diversity in the culture various notions are present for music signal. One thing that exists common among them is scale which can be vaguely defined as a set of pitches that repeat itself at a regular interval of time. These signals are meant for the pleasure of human listeners, and thus its features reflect specific aspects of human auditory perception. In particular, humans perceive two signals whose fundamental frequencies fall in a ratio 2:1 (an octave) as highly similar [2] (sometimes known as “octave equivalence”). Contemporary western music is rested on the “equal tempered” scale, that follows the mathematical coincidence whereby allowing the octave to be divided into twelve equal steps on a logarithmic axis while still preserving intervals corresponding to the most pleasant note combinations. These twelve step also known as the “white notes” on a piano, denoted by C, D, E, F, G, A, B. The lowest note on a piano is A0 (27.5 Hz), the highest note is C8 (4186 Hz), and middle C (262 Hz) is C4 [3]. Now if the signal is composed in accordance of the above method we could reside on concept of fundamental frequency or F0, which are an essential descriptor of harmonic sound signals such as speech and music. Single-F0 estimation algorithms assume that there is at most one harmonic source of which the F0 is to be extracted easily [4]. Music signal is now just limited to the pleasure objective. There many area where its inherit part of the things. Music pieces are taken for authentication purpose[5].Also with the era of digitalization the craze for video game has also increased the use of background music is also there[6]. This paper is to bridge the gap between the sparse representation and its effective utilization. Lots of work has been done for obtaining the sparse representation as it has due advantage the large amount of data turns to be non zero. This reduces the burden on the transmitting end and also the storage requirement. With the single-F0 estimation the presence of harmonic is detected .This can be done with help Fourier Frequency Transform (FFT) or Discrete Fourier Transform (DFT) for discrete data. The single-F0 estimation algorithms have developed. Its applications towards music signals are somehow limited because most music signals contain several concurrent harmonic. Processing of music can also be done in a symbolic framework; most commonly applied is the musical instrument digital interface (MIDI) as the input format. This kind of format exhibits several advantages over audio, since it is based on a considerably reduced amount of data, while incorporating much higher- level information in the form of note events and orchestration. However, the main limitation is that it loses some fine information available in audio signals such as frequency, amplitude modulations and spectral envelopes, which may be valuable for other tasks. Next is a mid-level representation based approach where the signal is decomposed into a small number of sound atoms or molecules bearing explicit musical instrument labels. Each atom is a sum of windowed harmonic sinusoidal partials whose relative amplitudes are specific to one instrument. The standard basis is also one of the most common approaches. Where, discrete wavelet transform or a wavelet packet is analyzed. The sparseness is a measure of how fast the DWT coefficients decay, and we are interested in obtaining a representation where the energy of the signal is concentrated in a few of the DWT coefficients. [7] Organization of the paper has been done in the following manner. It begins with the introduction in section I where a brief overview of the music and mainly concentrating on the western music scale as the signal opted for the analysis belongs to that class. This is followed by the work done to obtain the sparse representation of the music signal. Section II