Speaker Verification Using End-To-End Adversarial Language Adaptation

In this paper we investigate the use of adversarial domainadaptation for addressing the problem of language mismatchbetween speaker recognition corpora. In the context ofspeaker verification, adversarial domain adaptation methodsaim at minimizing certain divergences between the distributionthat the utterance-level features follow (i.e. speakerembeddings) when drawn from source and target domains(i.e. languages), while preserving their capacity in recognizingspeakers. Neural architectures for extracting utterancelevelrepresentations enable us to apply adversarial adaptationmethods in an end-to-end fashion and train the networkjointly with the standard cross-entropy loss. We examineseveral configurations, such as the use of (pseudo-)labels onthe target domain as well as domain labels in the feature extractor,and we demonstrate the effectiveness of our methodon the challenging NIST SRE16 and SRE18 benchmarks.