Speech is the most efficient and convenient way of communication. The learning capabilities of the deep learning architecture can be used to develop the sound classification system to overcome the efficiency issues of the traditional systems. We propose to develop a model that classifies the audio of the speaker.