Multi-Channel Far-Field Speaker Verification with Large-Scale Ad-hoc Microphone Arrays

TL;DR

This paper introduces a novel approach for speaker verification using large-scale ad-hoc microphone arrays, leveraging frame-level aggregation with attention mechanisms to exploit spatial-temporal information for improved accuracy.

Contribution

It proposes a new method that integrates cross-channel and cross-frame attention mechanisms, including graph attention, to enhance speaker verification performance in adverse environments.

Findings

Proposed methods achieve state-of-the-art results.

Graph attention outperforms self-attention in most cases.

Frame-level aggregation improves spatial-temporal information utilization.

Abstract

Speaker verification based on ad-hoc microphone arrays has the potential of reducing the error significantly in adverse acoustic environments. However, existing approaches extract utterance-level speaker embeddings from each channel of an ad-hoc microphone array, which does not consider fully the spatial-temporal information across the devices. In this paper, we propose to aggregate the multichannel signals of the ad-hoc microphone array at the frame-level by exploring the cross-channel information deeply with two attention mechanisms. The first one is a self-attention method. It consists of a cross-frame self-attention layer and a cross-channel self-attention layer successively, both working at the frame level. The second one learns the cross-frame and cross-channel information via two graph attention layers. Experimental results demonstrate that the proposed methods reach the state-of-the-art performance. Moreover, the graph-attention method is better than the self-attention method in most cases.