Speaker Verification By Partial AUC Optimization With Mahalanobis Distance Metric Learning

Abstract

Receiver operating characteristic (ROC) and detection error tradeoff (DET) curves are two widely used evaluation metrics for speaker verification. They are equivalent since the latter can be obtained by transforming the former's true positive y-axis to false negative y-axis and then re-scaling both axes by a probit operator. Real-world speaker verification systems, however, usually work on part of the ROC curve instead of the entire ROC curve given an application. Therefore, we propose in this paper to use the area under part of the ROC curve (pAUC) as a more efficient evaluation metric for speaker verification. A Mahalanobis distance metric learning based back-end is applied to optimize pAUC, where the Mahalanobis distance metric learning guarantees that the optimization objective of the back-end is a convex one so that the global optimum solution is achievable. To improve the performance of the state-of-the-art speaker verification systems by the proposed back-end, we further propose

Stats

• citations20
• S2 citations—
• github stars0
• HF likes0
• heat score9.92
• arxiv keybai2019speaker

Related papers

• Partial AUC Optimization Based Deep Speaker Embeddings With Class-center Learning For Text-independent Speaker Verification (2019)9.59
• Optimization Of The Area Under The ROC Curve Using Neural Network Supervectors For Text-dependent Speaker Verification (2019)10.21
• Multiobjective Optimization Training Of PLDA For Speaker Verification (2018)2.26
• A Speaker Verification Backend For Improved Calibration Performance Across Varying Conditions (2020)6.77
• Attention Back-end For Automatic Speaker Verification With Multiple Enrollment Utterances (2021)10.21
• A Discriminative Condition-aware Backend For Speaker Verification (2019)6.34
• Angular Softmax Loss For End-to-end Speaker Verification (2018)11.19
• A Comparison Of Metric Learning Loss Functions For End-to-end Speaker Verification (2020)6.77