Audio and video bit-rate reduction

The problem of estimating limits of bit-rate reduction for audio and video information is addressed. It is noted that binary encoding of such signals is performed by a variety of techniques, which are classified here into two categories (waveform and parameter). Encoders are compared on the basis of bits per Nyquist sample. Since minimizing the Nyquist sample requires a use of source signal statistics and human perception, audio and video sources and their statistics are discussed, as are the salient properties of hearing and vision and subjective assessments of quality. The present state of waveform and parameter encoding is assessed. It is predicted that waveform encoders will be refined to the point where bit rates will be nearly optimum but also highly variable. It is concluded that dramatic reductions will come about only through sophisticated parameter encoding.