Squeezing value of cross-domain labels: a decoupled scoring approach for speaker verification
This addresses domain mismatch issues in speaker verification systems, offering a principled solution for real-world applications where enrollment and test conditions differ.
The paper tackled the problem of domain mismatch in speaker verification, which causes performance reduction, by proposing a decoupled scoring approach that maximizes the value of cross-domain labels, achieving optimal verification scores in mismatched conditions.
Domain mismatch often occurs in real applications and causes serious performance reduction on speaker verification systems. The common wisdom is to collect cross-domain data and train a multi-domain PLDA model, with the hope to learn a domain-independent speaker subspace. In this paper, we firstly present an empirical study to show that simply adding cross-domain data does not help performance in conditions with enrollment-test mismatch. Careful analysis shows that this striking result is caused by the incoherent statistics between the enrollment and test conditions. Based on this analysis, we present a decoupled scoring approach that can maximally squeeze the value of cross-domain labels and obtain optimal verification scores when the enrollment and test are mismatched. When the statistics are coherent, the new formulation falls back to the conventional PLDA. Experimental results on cross-channel test show that the proposed approach is highly effective and is a principle solution to domain mismatch.