A multiresolution model of auditory excitation scalogram and its application to audio signal processing

This paper proposes a multiresolution model of auditory excitation scalogram and applies it to the problem of audio signal transform and coding techniques. The model uses wavelet packet (WP) transform for time-frequency decomposition of the input signal. The WP tree structure selection is based on an optimality criterion formulated to minimize entropy based cost functions. From applications' point of view, multiresolution model of the auditory excitation scalogram based on adaptive to the signal WP is used to achieve low bit rate coding of digital audio signals with minimum perceived loss of quality in reconstructed signal.