Music has been associate inherent a part of human life once it involves recreation; entertainment and far recently, even as a therapeutic medium. The means by which music is composed and listened to has witnessed a huge transition from the age of magnetic tape recorders to the recent age of digital music players streaming music from the cloud. What has remained intact is that the special relation that music shares with human emotions. We most frequently like better to hear a song or music which best suits our mood at that moment. In spite of this robust correlation, most of the music software existing nowadays are still barren of providing the ability of mood-aware play-list generation. This will increase the time music listeners soak in manually selecting category of songs fitting mood or occasion, which can be avoided by expanding upon songs with the relevant feeling class they convey. The downside, however, lies within the overhead of manual annotation of music with its corresponding mood and also the challenge is to spot this side mechanically and showing intelligence.We have analyzed many classification algorithms in order to be train and check the moods of audio songs.