DNN-Based Online Source Counting Based on Spatial Generalized Magnitude Squared Coherence

TL;DR

This paper introduces an online source counting method using spatial coherence and neural networks to detect changes in active sound sources in reverberant environments.

Contribution

It presents a novel approach combining spatial whitening, generalized magnitude-squared coherence, and neural networks for real-time source counting.

Findings

Effective in reverberant scenes with up to 4 speakers

Accurate detection of source count changes in real-time

Robust to background noise and reverberation

Abstract

The number of active sound sources is a key parameter in many acoustic signal processing tasks, such as source localization, source separation, and multi-microphone speech enhancement. This paper proposes a novel method for online source counting by detecting changes in the number of active sources based on spatial coherence. The proposed method exploits the fact that a single coherent source in spatially white background noise yields high spatial coherence, whereas only noise results in low spatial coherence. By applying a spatial whitening operation, the source counting problem is reformulated as a change detection task, aiming to identify the time frames when the number of active sources changes. The method leverages the generalized magnitude-squared coherence as a measure to quantify spatial coherence, providing features for a compact neural network trained to detect source count changes framewise. Simulation results with binaural hearing aids in reverberant acoustic scenes with up to 4 speakers and background noise demonstrate the effectiveness of the proposed method for online source counting.