Adaptation of Psychoacoustic analysis to wavelet domain in lossy audio coding

Abstract

The paper presents the authors' current research in adaptation of the MPEG AAC Psychoacoustic Model to the Discrete Wavelet Packet coding algorithm. We discovered the complex distribution of wavelet quantization noise across the audio spectrum due to significant cross-band aliasing. We also obtained the wavelet quantization noise function related to the analysis bands of MPEG AAC Psychoacoustic Model. The described approach provided the quantization matrix, which can be used for exact estimation of overall quantization noise during lossy coding.