The field of Lexical Semantic Change Detection (LSCD) deals with the detection of words that change their meaning over time. While there is a large amount of research in the field, only few go beyond a standard benchmark evaluation of existing models. The goal of this thesis is to derive a practical benefit from previous milestones in research. Therefore, a framework is built, that utilizes common approaches for LSCD to discover novel changing words. The framework is highly automated and easily applicable, making it useful for both beginners and experienced users. Anyone, who has access to two corpora (e.g., from different time periods) can use this framework to automatically discover words that change their meaning between the two corpora. In an exemplary discovery process, which includes multiple fine-tuning phases on common tasks, the framework and its underlying discovery process are demonstrated. The framework is successfully used to discover changing words between two time-specific German corpora. Additionally, in the fine-tuning phases the framework is also used to evaluate and optimize the implemented approaches and model parameters. The results show that the framework provided in this thesis and its implemented approaches can be used for the discovery of novel changing words and also evaluation.