Applying Auditory Masking to Speech Coding: Progress and Pitfalls (NOT KNOWN IF PUBLISHED BECAUSE AUTHOR HAS LEFT AT&T)

INTRODUCTION: As digital coding of speech signals moves to lower bit rates, it is no longer adequate to capitalize only on the redundancy inherent in the signal. To achieve high quality, speech coders operating at bit rates of 8 kbps and lower must make use of the properties of the signal source and of its ultimate destination - the human ear. One particularly relevant characteristic of auditory perception is that some signals can be made inaudible, or masked, by the presence of other signals.