在人工智慧引領的潮流下，服務型機器人日益普及我們生活當中，機器人的人機互動成為一項重要研究議題。對於國人而言使用中文語音實現人機互動是最具有親合力的方式，但是傳統的語音聲控技術是以固定語句轉換為單一動作的模式，除了缺乏使用彈性，系統的維護和擴充也很困難。本研究設計一個基於自然語言處理技術的中文語音控制系統，並以六軸機械手臂控制作為應用。此一系統使用了自動語音識別、中文分詞和句法解析等中介軟體，結合本文提出的參照樹架構來實現六軸機械手臂的中文語音控制。實驗結果印證本系統能以不同語者使用不同句構但具有相同語義的語音命令，使機械手臂做出正確對應語義的動作。我們同時提供一個語音命令擴充的框架，讓系統具備不同應用間的可移植性，以及可滿足使用者的個人化命令設定。;Under the trend of artificial intelligence, service robots are becoming more and more popular in our lives, and human-computer interaction of robots has become an important research topic. For Chinese people, using Chinese speech to realize human-computer interaction is the most familiar way. The traditional voice command device is a module to convert fixed statements into single actions. It’s lack the flexibility for daily used and also difficult for system maintenance and expansion. Based on natural language processing technology, it designed a Chinese speech control system with an application of 6-DOF robot arm in this research. This system used API such as automatic speech recognition, Chinese word segmentation and syntactic parsing and combined with the reference tree structure which proposed in this research to realize 6-DOF robot arm with Chinese speech control system. The experimental results show that the system can use different speech and same meaning sentences with different speaker to make the robot arm execute the correct action. We also provide a framework for voice command extensions, allowing the system to have portability between different applications and to meet the user′s personalized command settings.
