Manipulative and misleading news have become a commodity for some online news\noutlets and these news have gained a significant impact on the global mindset\nof people. Propaganda is a frequently employed manipulation method having as\ngoal to influence readers by spreading ideas meant to distort or manipulate\ntheir opinions. This paper describes our participation in the SemEval-2020,\nTask 11: Detection of Propaganda Techniques in News Articles competition. Our\napproach considers specializing a pre-trained BERT model on propagandistic and\nhyperpartisan news articles, enabling it to create more adequate\nrepresentations for the two subtasks, namely propaganda Span Identification\n(SI) and propaganda Technique Classification (TC). Our proposed system achieved\na F1-score of 46.060% in subtask SI, ranking 5th in the leaderboard from 36\nteams and a micro-averaged F1 score of 54.302% for subtask TC, ranking 19th\nfrom 32 teams.\n