Enhancement by postfiltering for speech and audio coding in ad-hoc sensor networks
This work addresses enhancement for speech and audio coding in ad-hoc sensor networks, but it is incremental as it builds on existing spatial filtering approaches by refining noise modeling.
The paper tackled the problem of signal quality degradation in wireless acoustic sensor networks due to quantization noise at low bitrates by proposing a Bayesian postfilter that explicitly models this noise, resulting in improved performance as measured by PSNR, PESQ, and MUSHRA scores.
Enhancement algorithms for wireless acoustics sensor networks~(WASNs) are indispensable with the increasing availability and usage of connected devices with microphones. Conventional spatial filtering approaches for enhancement in WASNs approximate quantization noise with an additive Gaussian distribution, which limits performance due to the non-linear nature of quantization noise at lower bitrates. In this work, we propose a postfilter for enhancement based on Bayesian statistics to obtain a multidevice signal estimate, which explicitly models the quantization noise. Our experiments using PSNR, PESQ and MUSHRA scores demonstrate that the proposed postfilter can be used to enhance signal quality in ad-hoc sensor networks.