Modelling Effective Algorithm for Media Fingerprinting Classification

Media fingerprinting classification must address the key problems like representation, search and etc. We propose an analysis of music information retrieval system based on audio clip sample time (CST). Suppose a query of music clip is received by a user through a cellular phone. This clip is processed before finding the similarity within the audio database. It is then matched with the database which is already available in music server. The proposed system has been exhaustively tested using different audio clip sample time and it shows about audio matching percentage (MP) is in the range of 68-75.17% when sec and 96.61-98% when sec. Here is the full length of the music file. Our results are very encouraging as our expectations. For noise-free environment these results show that the success rate of our system for music matching percentage versus query sample time are in the range of 68-75.17% (for sec) and 96.61-98% ( for sec). Here in sec is the full length of the music file. For noise-based environment these success rates of the presented system are also acceptable and up to the mark of expectations.