SJTU-AISPEECH System for VoxCeleb Speaker Recognition Challenge 2022

• DOI: 10.48550/arxiv.2209.09076
• Publication Date: 2022-01-01
• Authors: Chen, Zhengyang; Han, Bing; Xiang, Xu; Huang, Houjun; Liu, Bei; Qian, Yanmin
• Keywords: FOS: Computer and information sciences; Sound (cs.SD); Audio and Speech Processing (eess.AS); 0202 electrical engineering, electronic engineering, information engineering; FOS: Electrical engineering, electronic engineering, information engineering; 02 engineering and technology; Computer Science - Sound; Electrical Engineering and Systems Science - Audio and Speech Processing

This report describes the SJTU-AISPEECH system for the Voxceleb Speaker Recognition Challenge 2022. For track1, we implemented two kinds of systems, the online system and the offline system. Different ResNet-based backbones and loss functions are explored. Our final fusion system achieved 3rd place in track1. For track3, we implemented statistic adaptation and jointly training based domain adaptation. In the jointly training based domain adaptation, we jointly trained the source and target domain dataset with different training objectives to do the domain adaptation. We explored two different training objectives for target domain data, self-supervised learning based angular proto-typical loss and semi-supervised learning based classification loss with estimated pseudo labels. Besides, we used the dynamic loss-gate and label correction (DLG-LC) strategy to improve the quality of pseudo labels when the target domain objective is a classification loss. Our final fusion system achieved 4th place (very close to 3rd place, relatively less than 1%) in track3.<br/>System description of VoxSRC 2022<br/>