Subband coding of speech and audio

We discuss here the use of multirate filter banks for perceptually-weighted speech and audio compression. While our primary focus is on audio compression, we also review two recently proposed wideband speech coders that use filter banks to eliminate perceptually redundant information. Our goal is to examine and compare the time-frequency tradeoffs inherent in various coding algorithms as they try to exploit the masking properties of the human auditory system.