Online meetings often produce large volumes of information, making it challenging to extract important insights without getting lost in irrelevant details. To address this, the integration of large language models (LLMs) with speech recognition tools can help in generating concise summaries and minutes of such meetings. Effective summarization of virtual meetings relies heavily on crafting well-designed prompts that clearly instruct the LLM on what to summarize and how. Key elements of successful prompts include clarity, specificity, context, desired level of detail, language style, and inclusion of relevant questions. By incorporating these elements into prompts, users can enhance the productivity and effectiveness of AI-generated summaries, making it easier to derive meaningful insights from the extensive information discussed during meetings.