The goal of this report is to reproduce the experiment setup and verify the outcomes and conclusions of the presented paper called “Topic Modeling on Podcast Short-Text Metadata”. The scientific paper is based on podcasts and their genres, themes, and topics. The objective is to specify all kinds of podcasts only based on their metadata, without transcribed speech being considered. As podcasts are becoming massively consumed online content, listened for educational, entertainment, or informational purposes, and counting over 48 million episodes of podcasts worldwide, it is very important to be able to categorize and access information about these collections in the most effective way possible. It also proposes a new strategy to leverage named entities (NEs), often present in podcast metadata, in a Non-negative Matrix Factorization (NMF) topic modeling framework.