: Cover Song recognition from audio signals is a recent research topic in Human-Computer Interaction. Here we developed a system to detect cover from the audio signals. Cover Song Detection is created by extracting various audio features and combining them differently to create feature vectors. The feature vector is created using audio features like MFCC, MEL, and Chroma. Further, we also worked on using single, two and three combination of features. The cognizing was performed through Support Vector Machine (SVM), Naïve Bayes and K-Nearest Neighbor (KNN) along with employing the CNN. Importantly, CNN has shown notable improvement. The Proposed classifiers are compared with each other with combination of features. Their performances are compared using different classifier. Final proposed system works to identify best features and classifier for cover song detection.