Speaker Recognition Using Isomorphic Graph Attention Network Based Pooling on Self-Supervised Representation

TL;DR

This paper introduces IsoGAT, a novel graph attention pooling method for speaker recognition that leverages self-supervised speech representations, improving aggregation and recognition accuracy over existing methods.

Contribution

It proposes IsoGAT, an isomorphic graph attention network, for more effective pooling of self-supervised speech representations in speaker recognition.

Findings

IsoGAT outperforms existing pooling methods on VoxCeleb datasets.

The approach enhances speaker recognition accuracy using self-supervised features.

Experimental results validate the effectiveness of IsoGAT in real-world scenarios.

Abstract

The emergence of self-supervised representation (i.e., wav2vec 2.0) allows speaker-recognition approaches to process spoken signals through foundation models built on speech data. Nevertheless, effective fusion on the representation requires further investigating, due to the inclusion of fixed or sub-optimal temporal pooling strategies. Despite of improved strategies considering graph learning and graph attention factors, non-injective aggregation still exists in the approaches, which may influence the performance for speaker recognition. In this regard, we propose a speaker recognition approach using Isomorphic Graph ATtention network (IsoGAT) on self-supervised representation. The proposed approach contains three modules of representation learning, graph attention, and aggregation, jointly considering learning on the self-supervised representation and the IsoGAT. Then, we perform experiments for speaker recognition tasks on VoxCeleb1\&2 datasets, with the corresponding experimental results demonstrating the recognition performance for the proposed approach, compared with existing pooling approaches on the self-supervised representation.