White Noise Removal to Enhance Clarity of Sports Commentary

The spectators of any match make large amount of noise, which is captured along with live commentary during any telecast. The removal of noise created in spectator watching any sports event will improve the audibility of the commentary. Person always faces difficulty in perceiving multiple sounds coming from different sources simultaneously, known as the Cocktail Party problem. Removing the audience noise to improve the commentary quality is the core objective of this paper. Using the signal filter functionalities provided by MATLAB and its plugins, a solution is proposed. Analysis is done by collecting the sound from different sporting events. The case studies considered in the study vary in amount of spectators and the proposed approach produces a better solution for passive sports compared to active sports.