Psychoacoustically-based processing of MPEG-I layer 1-2 encoded signals

The MPEG audio standard has recently gained popularity for compressing high quality audio signals. In general, modifying or mixing MPEG signals requires decoding the MPEG signals into time-domain signals, processing the time-domain signals, and re-encoding the processed signals. In this paper, we present a psychoacoustically-based method for processing MPEG signals that eliminates significant portions of the encoding and decoding processes, resulting in reduced computational cost while maintaining perceptual audio quality