In this paper we propose and analyse a large margin fine-tuning strategy and a quality-aware score calibration in text-independent speaker verification. Large margin fine-tuning is a secondary training stage for DNN based speaker verification systems trained with margin-based loss functions. It enables the network to create more robust speaker embeddings by enabling the use of longer training utterances in combination with a more aggressive margin penalty. Score calibration is a common practice in speaker verification systems to map output scores to well-calibrated log-likelihood-ratios, which can be converted to interpretable probabilities. By including quality features in the calibration system, the decision thresholds of the evaluation metrics become quality-dependent and more consistent across varying trial conditions. Applying both enhancements on the ECAPA-TDNN architecture leads to state-of-the-art results on all publicly available VoxCeleb1 test sets and contributed to our winning submissions in the supervised verification tracks of the VoxCeleb Speaker Recognition Challenge 2020.
翻译:在本文中,我们提议并分析一个大型边际微调战略,并分析文本独立演讲者核查的质量比分校准。大边微调是DNN的演讲者核查系统接受基于边际损失功能培训的第二培训阶段。它使网络能够利用较长的培训话语权,加上更积极的边际罚款,从而建立更强有力的演讲者嵌入。评分校准是发言者核查制度中常见的做法,目的是将输出分数映射到校准的log-lihood-ratio,可以转换为可解释的概率。通过在校准系统中添加质量特征,评价指标的决定阈值就变得依赖质量,在不同试验条件下更加一致。对ECAPA-TDNN的两种改进都导致所有公开提供的VoxCeleb1测试组取得最新结果,并帮助我们在VoxCeleb议长承认挑战2020年监督核查轨道上获胜。