Subband audio coding with synthesis filters minimizing a perceptual distortion

The design of filter banks for source coding purposes classically relies on the perfect reconstruction (PR) property. However, several studies have shown that taking the quantization noise into account in the design could yield a noticeable reduction of the mean square reconstruction error. The purpose of this study is to show that perceptual improvement can also be obtained in the particular audio coding context by relaxing the PR constraint. In this context, the mean square error is not relevant any more, and we define a new perceptual distortion criterion, making use of a simplified ear model, the MPE (mean perceptual error). Then, synthesis filters are optimized so as to minimize this MPE. Finally, this MMPE (minimum MPE) filter bank is included in an audio coding scheme. Compared to the corresponding PR filter bank-based scheme by the means of POM (perceptual objective measure), they show an improved audio quality.

[1]  Thomas Sporer,et al.  -NMR- and -Masking Flag-: Evaluation of Quality Using Perceptual Criteria , 1992 .

[2]  Pierrick Philippe,et al.  Optimal wavelet packets for low-delay audio coding , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[3]  Pierre Duhamel,et al.  Filter bank design for minimum distortion in presence of subband quantization , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[4]  P. Duhamel,et al.  Modulated filter banks with minimum output distortion in presence of subband quantization , 1996, Conference Record of The Thirtieth Asilomar Conference on Signals, Systems and Computers.

[5]  P. P. Vaidyanathan,et al.  Statistically optimal synthesis banks for subband coders , 1994, Proceedings of 1994 28th Asilomar Conference on Signals, Systems and Computers.

[6]  Pierrick Philippe,et al.  A Musican-Like VQ Approach Yields Improved Low Bit-Rate Coding of Audio Signals , 1994 .

[7]  Richard A. Haddad,et al.  Modeling, analysis, and optimum design of quantized M-band filter banks , 1995, IEEE Trans. Signal Process..

[8]  Gérard Faucon,et al.  A Perceptual Objective Measurement System (POM) for the Quality Assessment of Perceptual Codecs , 1994 .

[9]  Pierre Duhamel,et al.  Perfect reconstruction versus MMSE filter banks in source coding , 1997, IEEE Trans. Signal Process..

[10]  Bor-Sen Chen,et al.  Optimal signal reconstruction in noisy filter bank systems: multirate Kalman synthesis filtering approach , 1995, IEEE Trans. Signal Process..

[11]  Stefanos D. Kollias,et al.  Optimal filter banks for signal reconstruction from noisy subband components , 1996, IEEE Trans. Signal Process..

[12]  Jean-Bernard Rault,et al.  A New Noise Injection Model for Audio Compression Algorithms , 1996 .

[13]  Jelena Kovacevic,et al.  Subband coding systems incorporating quantizer models , 1995, IEEE Trans. Image Process..