A method and system for automatic detection of the end of a video stream, wherein the energy level of the audio content is computed, a window at a time. The computed audio energy level is then compared against one or more audio energy level thresholds. Various features of the color content of the video content are then computed, a frame at a time. The computed color content is then compared against one or more video thresholds. Based on the results of the audio and video threshold comparisons, the end of the video stream can be automatically detected.