Audio Spotforming Using Nonnegative Tensor Factorization with Attractor-Based Regularization

TL;DR

This paper introduces a novel nonnegative tensor factorization approach with attractor-based regularization for improved target-speaker extraction in multi-microphone array systems, enhancing robustness and practical applicability.

Contribution

It proposes a new NTF-based common component extraction method with attractor regularization for more interpretable and robust spotforming.

Findings

Outperforms conventional methods in spotforming accuracy

Demonstrates robustness against hyperparameter variations

Shows potential for practical deployment

Abstract

Spotforming is a target-speaker extraction technique that uses multiple microphone arrays. This method applies beamforming (BF) to each microphone array, and the common components among the BF outputs are estimated as the target source. This study proposes a new common component extraction method based on nonnegative tensor factorization (NTF) for higher model interpretability and more robust spotforming against hyperparameters. Moreover, attractor-based regularization was introduced to facilitate the automatic selection of optimal target bases in the NTF. Experimental results show that the proposed method performs better than conventional methods in spotforming performance and also shows some characteristics suitable for practical use.