Enhancement by postfiltering for speech and audio coding in ad-hoc sensor networks

TL;DR

This paper introduces a Bayesian-based postfilter for wireless sensor networks that explicitly models quantization noise, significantly improving speech and audio quality at low bitrates.

Contribution

It presents a novel postfilter leveraging Bayesian statistics to better handle quantization noise in ad-hoc sensor networks, enhancing signal quality.

Findings

Improved PSNR, PESQ, and MUSHRA scores demonstrate effectiveness.

Explicit quantization noise modeling enhances enhancement performance.

Postfilter outperforms conventional spatial filtering approaches.

Abstract

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.