STUDY ON FRAMEWORK OF AUDIO FINGERPRINTING AND SHAZAM?S WORKING ALGORITHM

Inquisitive to know the songs and vanquishing the idea of remembering many songs lead to the innovation of this most anticipated music identification application called Shazam. Many music identification applications have been developed, but some of the most important features which are applied in Shazam makes it distinguishable from others. Shazam is an adaptable music quest engine and it identifies audio file based on the music played or recorded with the concept of fingerprinting. Fingerprint summarizes the whole audio file. Audio fingerprinting is used to match the song from database using limited number of bits. It?s computational time is quick and easy with few faulty positives and excessive identification rate. By using fingerprinting mechanism, there is no requirement of any details of the song like singer , movie name to identify the song. Fingerprint is extracted from a framework after progressing through specific stages. Shazam uses fingerprints to identify the song which is obtained by analog to digital conversion, sampling, quantizing ,pulse coded modulation ,down sampling .Mainly the working algorithm of shazam uses combinatory hashed time and frequency pattern or topological scrutiny of the audio and finally querying for match of hashes in the database. Shazam ignores the noise added to the music while recording when you try to identify a song in a noisy environment. Shazam is significantly expandable. These concepts will be divulged in this paper and gives the readers a fundamental overview of Shazam?s working algorithm.