Re-encoding of perceptually quantized wavelet packet transform coefficients of audio and high quality speech

This paper reports on the results of four re-encoding schemes on perceptually quantized wavelet packet transform (WPT) coefficients of audio and high quality speech. These schemes comprises: 1- embedded zero-tree wavelet (EZW) 2- The set partitioning in hierarchical trees (SPIHT) 3-JPEG-based entropy/run length Huffman and 4-JPEG-type audio Huffman coding algorithms. Since EZW and SPIHT are designed for image compression, some new modifications have been implemented in these schemes for their better matching with audio signals. The performances of these four re-encoders are compared in terms of average output bit rate and computation time of a same codec. It is concluded that the JPEG-type audio huffman coding achieves the best results although it is not possible to truncate the bit stream, in this case, to easily match the bit rate to the fixed channel capacity.