Virtual Reality Systems project, Software Engineering and Information Technologies, Master Academic Studies, Faculty of Technical Sciences, University of Novi Sad, 2020/2021
Datasets used: VGG Face, VoxCeleb.
The VGGFace model was used for face classification.
The VGGVox model was used for speaker recognition. The weights can be downloaded from here.
Due to the limitations of Google's free API, any video used must be no longer than 30 seconds.