We propose a new multiple description spherical quantization with repetitively coded amplitudes (MDSQRA) scheme suited for quantization of sinusoidal parameters. The quantization scheme is constituted by a set of spherical quantizers inspired by the multiple description spherical trellis-coded quantization (MDSTCQ) scheme. In this scheme, we apply repetitive coding on the amplitudes, while multiple description coding are applied on the phases and frequencies. Thereby, MDSQRA becomes directly implementable, as opposed to MDSTCQ, since the phase and frequency quantizers depend on the amplitudes which have dissimilar descriptions in MDSTCQ. Furthermore, we implement MDSQRA into a perceptual matching pursuit based sinusoidal audio coder. Finally, we evaluate MDSQRA through perceptual distortion measurements and MUSHRA listening tests. The tests show that MDSQRA outperforms MDSTCQ with respect to a expected perceptual distortion measure. The same results are obtained through the MUSHRA tests performed on sound clips coded using MDSQRA and MDSTCQ.Index Terms-Multiple description coding, perceptual audio coding, pre-and post-filtering, sinusoidal parametric coding, spherical quantization.

Recently, multiple description spherical trellis-coded quantization (MDSTCQ) for quantization of sinusoidal parameters was proposed, which suffered from a suboptimal implementation. Therefore, we propose the multiple description spherical quantization with repetition coding of the amplitudes (MD-SQRA) scheme. Here, we apply repetition coding to the amplitudes, whereas the phases and the frequencies are coded using multiple description quantization. We measure the relation between the expected perceptual distortions for the two quantization schemes, which shows that MDSQRA outperforms MDSTCQ for most packet-loss probabilities. Also, we measure the perceptual distortion obtained when applying MDSQRA on synthetic data and show that it matches the expected perceptual distortion. Furthermore, due to MD-STCQs suboptimal implementation, MDSQRA outperforms MDSTCQ in terms of measured perceptual distortion even at packet-loss probabilities where the expected perceptual distortion was lowest for MDSTCQ.