Extending GCC-PHAT using Shift Equivariant Neural Networks

TL;DR

This paper introduces a shift equivariant neural network extension to GCC-PHAT for improved speaker localization, maintaining theoretical guarantees in ideal conditions and enhancing robustness in noisy, reverberant environments.

Contribution

It proposes a novel neural network-based extension to GCC-PHAT that preserves timing information and improves performance in challenging acoustic conditions.

Findings

Reduces localization error in adverse environments

Maintains exact delay recovery in ideal conditions

Outperforms traditional GCC-PHAT in noisy settings

Abstract

Speaker localization using microphone arrays depends on accurate time delay estimation techniques. For decades, methods based on the generalized cross correlation with phase transform (GCC-PHAT) have been widely adopted for this purpose. Recently, the GCC-PHAT has also been used to provide input features to neural networks in order to remove the effects of noise and reverberation, but at the cost of losing theoretical guarantees in noise-free conditions. We propose a novel approach to extending the GCC-PHAT, where the received signals are filtered using a shift equivariant neural network that preserves the timing information contained in the signals. By extensive experiments we show that our model consistently reduces the error of the GCC-PHAT in adverse environments, with guarantees of exact time delay recovery in ideal conditions.