A two-step approach for speech enhancement in low-SNR scenarios using cyclostationary beamforming and DNNs

TL;DR

This paper introduces a two-step speech enhancement method combining cyclostationary spectral beamforming with lightweight DNNs, significantly improving noise suppression in low-SNR scenarios with harmonic noise.

Contribution

It proposes a cyclostationarity-aware preprocessing step using cMPDR beamforming that enhances DNN-based speech enhancement without modifying neural network architectures.

Findings

cMPDR preprocessing improves noise suppression at low SNRs

CRNN with cMPDR outperforms larger ULCNet on raw inputs

Explicit cyclostationarity modeling surpasses increasing model complexity

Abstract

Deep Neural Networks (DNNs) often struggle to suppress noise at low signal-to-noise ratios (SNRs). This paper addresses speech enhancement in scenarios dominated by harmonic noise and proposes a framework that integrates cyclostationarity-aware preprocessing with lightweight DNN-based denoising. A cyclic minimum power distortionless response (cMPDR) spectral beamformer is used as a preprocessing block. It exploits the spectral correlations of cyclostationary noise to suppress harmonic components prior to learning-based enhancement and does not require modifications to the DNN architecture. The proposed pipeline is evaluated in a single-channel setting using two DNN architectures: a simple and lightweight convolutional recurrent neural network (CRNN), and a state-of-the-art model, namely ultra-low complexity network (ULCNet). Experiments on synthetic data and real-world recordings dominated by rotating machinery noise demonstrate consistent improvements over end-to-end DNN baselines, particularly at low SNRs. Remarkably, a parameter-efficient CRNN with cMPDR preprocessing surpasses the performance of the larger ULCNet operating on raw or Wiener-filtered inputs. These results indicate that explicitly incorporating cyclostationarity as a signal prior is more effective than increasing model capacity alone for suppressing harmonic interference.