Description: The employment of digital wireless microphones in professional contexts requires ultra-low delay, strong robustness, and high audio quality. On the other hand, audio coding is required in order to comply with the restrictions on the bandwidth of the radio channel, making the resulting source-coded audio signal more vulnerable to channel distortions. Therefore, in this paper we present a transmission system for differential pulse-code modulated (DPCM) audio with receiver-sided soft-decision error concealment exploiting channel reliability information, explicit redundancy by simple delayless parity-check codes, and residual redundancy within the source-coded audio signal. Simulations on frequency-shift keying (FSK)-modulated channels with additive white Gaussian noise show considerable gains in audio quality compared to hard-decision decoding and soft-decision decoding only exploiting reliability information and 0th-order a priori knowledge.

Description: For high-quality digital audio transmitted over error-prone shortrange wireless channels robust source decoding with low decoding delay is desired. However, many previous approaches from the field of audio error concealment are solely intuitively motivated or introduce algorithmic delay. In contrast, this paper deals with a Bayesian framework for delayless full-band soft-decision error concealment for quantized but uncompressed audio utilizing only residual redundancy in the audio signal and channel reliability information. In principle, it can be applied to any channel providing reliability information. As a novelty, we employ multiple adaptively shaped prior probability distributions in the decoding process. These are utilized in conjunction with the autocorrelation method or the normalized least-mean-square (NLMS) algorithm in order to compute prediction probabilities within the Bayesian framework. Experiments carried out on representative audio data transmitted over additive white Gaussian noise (AWGN) channels show significant enhancements in audio quality.