The diagram above gives a graphical representation of our intended approach towards the use of computer generated images in conjunction with television. The main aim of this idea is to filter out the spoken words from the sound effects in television broadcasts, enabling visual representations of the two to be placed in two corners of the television screen. These "wave-like" images would represent the pitch and loudness of peoples voices in one corner, and similarly the magnitude and general waveform of the explosions and creaking doors in the other.