Detection and Analysis of Content Creator Collaborations in YouTube Videos using Face- and Speaker-Recognition

Abstract

This paper demonstrates the application of speaker recognition for the detection of collaborations in YouTube videos. Here, we extend CATANA, which is an existing framework for detection and analysis of YouTube collaborations that utilizes face recognition for the detection of collaborators and naturally performs poor on video-content without appearing faces. This work proposes an extension that uses active speaker detection and speaker recognition to improve the detection accuracy in YouTube collaborations.