This research will study the comparison of more than one real time BSS algorithm based on ICA theoretical. There are two main goals: (1) to obtain the fastest separation possible that works online with minimal delay by implementing and testing various algorithms., (2) Tto gain the knowledge of the advantages and disadvantages in order to derive an algorithm with the minimum delay value wthichat operates in real-time and outputs the cleanest speech. This research will measure the possibility of reducing the volume to obtain the lowest value of background music and the possibility of reducing the delay of speech to coincide with the show. Many algorithms will be tested and compared to choose the best algorithms in terms of (the lowest value of the delay of speech and the cleanest sound speech from music background).