Microsoft Speaker Diarization System for the VoxCeleb Speaker Recognition Challenge 2020
Abstract
This paper describes the Microsoft speaker diarization system for monaural multi-talker recordings in the wild, evaluated at the diarization track of the VoxCeleb Speaker Recognition Challenge(VoxSRC) 2020. We will first explain our system design to address issues in handling real multi-talker recordings. We then present the details of the components, which include Res2Net-based speaker embedding extractor, conformer-based continuous speech separation with leakage filtering, and a modified DOVER (short for Diarization Output Voting Error Reduction) method for system fusion. We evaluate the systems with the data set provided by VoxSRCchallenge 2020, which contains real-life multi-talker audio collected from YouTube. Our best system achieves 3.71% and 6.23% of the diarization error rate (DER) on development set and evaluation set, respectively, being ranked the 1st at the diarization track of the challenge.
Keywords
Cite
@article{arxiv.2010.11458,
title = {Microsoft Speaker Diarization System for the VoxCeleb Speaker Recognition Challenge 2020},
author = {Xiong Xiao and Naoyuki Kanda and Zhuo Chen and Tianyan Zhou and Takuya Yoshioka and Sanyuan Chen and Yong Zhao and Gang Liu and Yu Wu and Jian Wu and Shujie Liu and Jinyu Li and Yifan Gong},
journal= {arXiv preprint arXiv:2010.11458},
year = {2020}
}
Comments
5 pages, 3 figures, 2 tables