進行會議是人與人之間重要的交流活動，一個完整的會議除了包含進行訊息交換的雙方人員，最重要的是會議結束後的討論結果。常見的會議紀錄方式包含：手寫、錄音或錄影等，基本上都需要由人工方式來處理，而由人工處理必定需要額外花費多餘的人力及時間來保證其功效及正確性。智慧會議室的目標即在改變這種現況，並提升會議的效率。 本論文擬在「智慧會議室」的主題之內，建立一套「影像與語音整合系統」，使與會人員僅需透過按下麥克風發話鍵的方式，攝影鏡頭即可自動轉向，藉此減少人工時間，提升作業效率，並保障會議紀錄的正確性。 Meeting is very important between people and people with communication. The most important result of a meeting is the discussions during the meeting. Generally, the meeting records consist of handwriting, audio and/or video recordings. Nowadays, the recording still needs manually handling which takes a considerable extra time and extra manpower to ensure its effectiveness and correctness. The "Smart Conference Room" is a solution to improve this situation and enhance the efficiency of the meeting management. This thesis intends to develop a function of the "Smart Conference Room." We try to establish a "Voice-and-Video Integrated System." When the participants want to speak simply by pressing the microphone talking key, meanwhile, the developed system automatically control the camera automatically turn to the speaker such that the speaker can be the main focus of the camera. The work of this thesis can reduce manual handled time, improve the management efficiency of the meeting as well as guarantee the correctness of the meeting record.