I worked on synthetic voice spoofing detection using one-class learning for a university project.
I got interested in this project because of the Cyber Security element present in it.
This paper shows how one-class classification can be used to improve the detection model's performance.
This is the research paper I used as a reference for the project.
I summarised the project into a presentation.
You can also download the pptx file.
I was able to reproduce the experiment done for this project on my
GitHub