Neural Network-Based Time-Frequency-Bin-Wise Linear Combination of Beamformers for Underdetermined Target Source Extraction

TL;DR

This paper introduces a neural network framework for combining beamformers in underdetermined source extraction, improving temporal-spectral coherence and performance without needing noise covariance estimation.

Contribution

The proposed NN-TFLC method predicts coherent combination weights using a cross-attention mechanism, enhancing source extraction in underdetermined mixtures.

Findings

Outperforms traditional TFS/TFLC methods in experiments.

Achieves performance comparable to MVDR-based approaches without noise priors.

Provides a neural network approach for coherent beamformer combination.

Abstract

Extracting a target source from underdetermined mixtures is challenging for beamforming approaches. Recently proposed time-frequency-bin-wise switching (TFS) and linear combination (TFLC) strategies mitigate this by combining multiple beamformers in each time-frequency (TF) bin and choosing combination weights that minimize the output power. However, making this decision independently for each TF bin can weaken temporal-spectral coherence, causing discontinuities and consequently degrading extraction performance. In this paper, we propose a novel neural network-based time-frequency-bin-wise linear combination (NN-TFLC) framework that constructs minimum power distortionless response (MPDR) beamformers without explicit noise covariance estimation. The network encodes the mixture and beamformer outputs, and predicts temporally and spectrally coherent linear combination weights via a cross-attention mechanism. On dual-microphone mixtures with multiple interferers, NN-TFLC-MPDR consistently outperforms TFS/TFLC-MPDR and achieves competitive performance with TFS/TFLC built on the minimum variance distortionless response (MVDR) beamformers that require noise priors.