Extending GCC-PHAT using Shift Equivariant Neural Networks
This work addresses noise and reverberation issues in microphone array-based speaker localization, offering an incremental improvement over existing neural network-enhanced GCC-PHAT methods.
The paper tackles the problem of speaker localization in adverse acoustic environments by extending the GCC-PHAT method with a shift equivariant neural network, resulting in consistently reduced error while preserving exact time delay recovery in ideal conditions.
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.