Objectives: Raga's role in classical Indian music never ends. The music of the Indian subcontinent is known as Indian classical music. Raga recognition is crucial for automated Raga identification systems.Raga is a set of various distinctive notes (swara) with certain special characteristics. Methods:These method help us to define the specific raga as well as the musical instrument performed in a specified polyphonic audio signals.Despite earlier methods to this issue, which mainly rely on pitch features, this paper seek in this study to construct hybrid features that can learn the distinguishing traits of a raga from the prevailing pitch elements in a music. Findings: Here extraction of features is performed using spectral centroid, spectral crest factor, spectral decrease, spectral flatness, spectral flux features. Support Vector Machine (SVM) algorithm is use to classify the characteristics as the number of clusters in a specified dataset is estimated. Finally, the raga is labeled using a supervised learning algorithm called Support Vector Machine(SVM).