通过维度正则化与分数归一化拓展自蒸馏原型网络的前沿边界
Pushing the Frontiers of Self-Distillation Prototypes Network with Dimension Regularization and Score Normalization
AI总结：
本研究针对无标签说话人验证中自监督与监督方法的性能差距，通过为自蒸馏原型网络添加维度正则化解决坍缩问题、结合分数归一化技术，在VoxCeleb1基准上大幅刷新自监督说话人验证的最优性能。
摘要：
在无说话人标签的条件下开发鲁棒的说话人验证（SV）系统是一项长期存在的挑战。早期研究表明，自监督方法与全监督方法之间存在显著的性能差距。本文对非对比自监督框架Self-Distillation Prototypes Network（SDPN，自蒸馏原型网络）进行改进，引入维度正则化，通过对说话人嵌入施加正则项来明确解决坍缩问题。此外，我们整合了全监督说话人验证中的分数归一化技术，进一步缩小与监督验证性能的差距。加入维度正则化和分数归一化的SDPN在VoxCeleb1说话人验证评估基准上创下新的最优性能，在VoxCeleb1-{O,E,H}测试集上的等错误率分别达到1.29%、1.60%和2.80%。这些结果相较于当前最优的自监督方法分别实现了28.3%、19.6%和22.6%的相对提升，推动了说话人验证技术的前沿发展。
英文摘要：
Developing robust speaker verification (SV) systems without speaker labels has been a longstanding challenge. Earlier research has highlighted a considerable performance gap between self-supervised and fully supervised approaches. In this paper, we enhance the non-contrastive self-supervised framework, Self-Distillation Prototypes Network (SDPN), by introducing dimension regularization that explicitly addresses the collapse problem through the application of regularization terms to speaker embeddings. Moreover, we integrate score normalization techniques from fully supervised SV to further bridge the gap toward supervised verification performance. SDPN with dimension regularization and score normalization sets a new state-of-the-art on the VoxCeleb1 speaker verification evaluation benchmark, achieving Equal Error Rate 1.29%, 1.60%, and 2.80% for trial VoxCeleb1-{O,E,H} respectively. These results demonstrate relative improvements of 28.3%, 19.6%, and 22.6% over the current best self-supervised methods, thereby advancing the frontiers of SV technology.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
