The Vicomtech Spoofing-Aware Biometric System for the SASV Challenge

TL;DR

This paper presents a spoofing-aware speaker verification system that combines neural network embeddings with an integration network trained on a one-class loss, achieving competitive results on the ASVspoof19 database.

Contribution

The novel integration approach combines speaker verification and antispoofing embeddings with a one-class loss trained network for improved spoofing detection.

Findings

Competitive performance on ASVspoof19 database

Effective combination of verification and antispoofing embeddings

High-performance speech biometric systems with self-supervised learning

Abstract

This paper describes our proposed integration system for the spoofing-aware speaker verification challenge. It consists of a robust spoofing-aware verification system that use the speaker verification and antispoofing embeddings extracted from specialized neural networks. First, an integration network, fed with the test utterance's speaker verification and spoofing embeddings, is used to compute a spoof-based score. This score is then linearly combined with the cosine similarity between the speaker verification embeddings from the enrollment and test utterances, thus obtaining the final scoring decision. Moreover, the integration network is trained using a one-class loss function to discriminate between target trials and unauthorized accesses. Our proposed system is evaluated in the ASVspoof19 database, exhibiting competitive performance compared to other integration approaches. In addition, we test, along with our integration approach, state of the art speaker verification and antispoofing systems based on self-supervised learning, yielding high-performance speech biometric systems.