Sparsity-based audio declipping methods: selected overview, new algorithms, and large-scale evaluation

TL;DR

This paper reviews sparsity-based audio declipping methods, introduces new algorithms within a unified framework, and provides large-scale evaluations to guide practitioners in choosing suitable techniques based on clipping severity and audio content.

Contribution

It proposes a comprehensive algorithmic framework for declipping, compares various methods systematically, and emphasizes the need for larger benchmarks with perceptually significant clipping levels.

Findings

Larger-scale benchmarks reveal limitations of existing methods at high clipping levels.

Structured sparsity techniques outperform plain sparsity in declipping tasks.

Guidelines are provided for selecting declipping methods based on clipping severity and audio type.

Abstract

Recent advances in audio declipping have substantially improved the state of the art.% in certain saturation regimes. Yet, practitioners need guidelines to choose a method, and while existing benchmarks have been instrumental in advancing the field, larger-scale experiments are needed to guide such choices. First, we show that the clipping levels in existing small-scale benchmarks are moderate and call for benchmarks with more perceptually significant clipping levels. We then propose a general algorithmic framework for declipping that covers existing and new combinations of variants of state-of-the-art techniques exploiting time-frequency sparsity: synthesis vs. analysis sparsity, with plain or structured sparsity. Finally, we systematically compare these combinations and a selection of state-of-the-art methods. Using a large-scale numerical benchmark and a smaller scale formal listening test, we provide guidelines for various clipping levels, both for speech and various musical genres. The code is made publicly available for the purpose of reproducible research and benchmarking.