MPEG audio compression future

Abstract

We have seen that MPEG-1 and MPEG-2 Audio coders used perceptually shaped quantization noise as the primary tool for achieving compression. MPEG-4 High-Efficiency AAC [1] added parametric coding of the upper spectrum region with the Spectral Band Replication tool and parametric coding of two-channel signals with the Parametric Stereo tool. The MPEG-D standards MPEG Surround [2] and Spatial Audio Object Coding [3] incorporate parametric coding of the sound stage using level, time and coherence parameters for regions of the time/frequency signal. The common thread in all of these tools is that they model and exploit how humans perceive sound.