Business process modeling is a crucial task for organizations to document, analyze, and optimize their business processes. However, creating process models requires modeling expertise, aggravating this knowledge-intensive task for non-experts. Recent advancements in Natural Language Processing and artificial intelligence offer new possibilities for automating the generation of process models from natural language descriptions. This doctoral research aims to address these emerging possibilities by developing BPMNGen, an LLM-based conversational framework that enables users to generate and iteratively evolve BPMN 2.0 process models using natural language input. By leveraging NLP techniques, BPMNGen translates the process descriptions entered by the user into BPMN 2.0 process models and then allows for real-time modifications through interactive prompts. The research follows the Design Science Research Methodology to develop and evaluate the conversational framework in a structured manner. A key focus of future work will be the evaluation of the generated BPMN 2.0 process models to ensure their quality, correctness, and usefulness. Additionally, different techniques (including prompt engineering) will be explored to improve the accuracy and quality of the model generation process. The main goal of this research is to make process modeling with BPMN 2.0 more accessible and convenient for a broader audience, including domain experts, process participants, and business analysts.