Speaker Diarization with Session-Level Speaker Embedding Refinement Using Graph Neural Networks

被引：0

作者：

Wang, Jixuan ^{[1
]}

Xiao, Xiong ^{[2
]}

Wu, Jian ^{[2
]}

Ramamurthy, Ranjani ^{[2
]}

Rudzicz, Frank ^{[1
]}

Brudno, Michael ^{[1
]}

机构：

[1] University of Toronto, Canada

[2] Microsoft, United States

来源：

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | 2020年 / 2020-May卷

关键词：

Compendex;

D O I：

9054176

中图分类号：

学科分类号：

摘要：

Graph neural networks - Speech recognition - Deep neural networks - Clustering algorithms - Matrix algebra

引用

页码：7109 / 7113