Biodenoising: Animal Vocalization Denoising without Access to Clean Data

TL;DR

This paper introduces a novel animal vocalization denoising method that leverages pseudo-clean targets derived from diverse bioacoustics datasets, enabling effective denoising without access to large clean datasets.

Contribution

It proposes a new training approach using pseudo-clean targets and creates a diverse benchmark set for animal vocalization denoising evaluation.

Findings

Denoising models trained on pseudo-clean targets perform competitively on the benchmark.

The approach works across multiple species and acoustic environments.

The dataset and tools are publicly available for further research.

Abstract

Animal vocalization denoising is a task similar to human speech enhancement, which is relatively well-studied. In contrast to the latter, it comprises a higher diversity of sound production mechanisms and recording environments, and this higher diversity is a challenge for existing models. Adding to the challenge and in contrast to speech, we lack large and diverse datasets comprising clean vocalizations. As a solution we use as training data pseudo-clean targets, i.e. pre-denoised vocalizations, and segments of background noise without a vocalization. We propose a train set derived from bioacoustics datasets and repositories representing diverse species, acoustic environments, geographic regions. Additionally, we introduce a non-overlapping benchmark set comprising clean vocalizations from different taxa and noise samples. We show that that denoising models (demucs, CleanUNet) trained on pseudo-clean targets obtained with speech enhancement models achieve competitive results on the benchmarking set. We publish data, code, libraries, and demos at https://earthspecies.github.io/biodenoising/.