Inference-Adaptive Steering of Neural Networks for Real-Time Area-Based Sound Source Separation

Strauß M, Mack W, Valero ML, Kopuklu O (2025)

Publication Type: Journal article

Publication year: 2025

Journal

IEEE Signal Processing Letters Institute of Electrical and Electronics Engineers (IEEE)

DOI: 10.1109/LSP.2025.3543454

Abstract

We propose a novel adaptive steering technique that changes the target area of a spatial-aware multi-microphone sound source separation algorithm during inference without the necessity of retraining the deep neural network (DNN). To achieve this, we first train a DNN aiming to retain speech within a target region, defined by an angular span, while suppressing sound sources stemming from other directions. Afterward, a phase shift is applied to the microphone signals, allowing us to shift the center of the target area during inference at negligible additional cost in computational complexity. Further, we show that the proposed approach performs well in a wide variety of acoustic scenarios, including several speakers inside and outside the target area and additional noise. More precisely, the proposed approach performs on par with DNNs trained explicitly for the steered target area in terms of DNSMOS and SI-SDR.

Authors with CRIS profile

Martin Strauß Lehrstuhl für Audiosignalanalyse (AudioLabs) Wolfgang Mack International Audio Laboratories Erlangen (AudioLabs)

Involved external institutions

LMU Klinikum DE Germany (DE)

How to cite

APA:

Strauß, M., Mack, W., Valero, M.L., & Kopuklu, O. (2025). Inference-Adaptive Steering of Neural Networks for Real-Time Area-Based Sound Source Separation. IEEE Signal Processing Letters. https://doi.org/10.1109/LSP.2025.3543454

MLA:

Strauß, Martin, et al. "Inference-Adaptive Steering of Neural Networks for Real-Time Area-Based Sound Source Separation." IEEE Signal Processing Letters (2025).

BibTeX: Download