Semi-Supervised Sound Source Localization Based on Manifold Regularization

TL;DR

This paper introduces a semi-supervised sound source localization method that leverages manifold regularization to improve accuracy in challenging acoustic environments, using minimal labeled data and adapting with new unlabelled samples.

Contribution

It proposes a novel manifold regularization-based algorithm for source localization that is adaptive and effective with limited labeled data.

Findings

Outperforms existing manifold learning-based localization algorithms

Achieves higher accuracy than generalized cross-correlation baseline

Demonstrates robustness in reverberant and low SNR conditions

Abstract

Conventional speaker localization algorithms, based merely on the received microphone signals, are often sensitive to adverse conditions, such as: high reverberation or low signal to noise ratio (SNR). In some scenarios, e.g. in meeting rooms or cars, it can be assumed that the source position is confined to a predefined area, and the acoustic parameters of the environment are approximately fixed. Such scenarios give rise to the assumption that the acoustic samples from the region of interest have a distinct geometrical structure. In this paper, we show that the high dimensional acoustic samples indeed lie on a low dimensional manifold and can be embedded into a low dimensional space. Motivated by this result, we propose a semi-supervised source localization algorithm which recovers the inverse mapping between the acoustic samples and their corresponding locations. The idea is to use an optimization framework based on manifold regularization, that involves smoothness constraints of possible solutions with respect to the manifold. The proposed algorithm, termed Manifold Regularization for Localization (MRL), is implemented in an adaptive manner. The initialization is conducted with only few labelled samples attached with their respective source locations, and then the system is gradually adapted as new unlabelled samples (with unknown source locations) are received. Experimental results show superior localization performance when compared with a recently presented algorithm based on a manifold learning approach and with the generalized cross-correlation (GCC) algorithm as a baseline.