Position tracking of a varying number of sound sources with sliding permutation invariant training
This addresses the challenge of consistent multi-source tracking in real-world acoustic environments, representing an incremental improvement over existing methods.
The paper tackled the problem of tracking a varying number of moving sound sources in acoustic scenarios by introducing a new training strategy for deep learning sound source localization models, resulting in reduced identity switches without compromising localization accuracy.
Recent data- and learning-based sound source localization (SSL) methods have shown strong performance in challenging acoustic scenarios. However, little work has been done on adapting such methods to track consistently multiple sources appearing and disappearing, as would occur in reality. In this paper, we present a new training strategy for deep learning SSL models with a straightforward implementation based on the mean squared error of the optimal association between estimated and reference positions in the preceding time frames. It optimizes the desired properties of a tracking system: handling a time-varying number of sources and ordering localization estimates according to their trajectories, minimizing identity switches (IDSs). Evaluation on simulated data of multiple reverberant moving sources and on two model architectures proves its effectiveness on reducing identity switches without compromising frame-wise localization accuracy.