BriefGPT.xyz
Jul, 2022
无监督说话人分离：对语言，重叠以及参数调整不敏感
Unsupervised Speaker Diarization that is Agnostic to Language, Overlap-Aware, and Tuning Free
HTML
PDF
M. Iftekhar Tanveer, Diego Casabuena, Jussi Karlgren, Rosie Jones
TL;DR
本文探讨了一种不依赖于语种特定组件的无监督的演讲者分离技术，该算法具有覆盖感知功能，不需要有关演讲者数量的信息， 并在播客数据上表现出了79％的纯度得分改进（34％的F-分数）。
Abstract
podcasts
are conversational in nature and speaker changes are frequent -- requiring
speaker diarization
for
content understanding
. We prop
→