日本語文法学習支援システムの開発──音声認識と音声合成などの自然言語処理技術を利用した例にして

The Development of a Japanese Grammar Learning Support System-The Case of Using Natural Language Processing Technologies of Speech Recognition and Voice Synthesis

最近幾年高速網路的利用變得更加普及，以網路為基礎的多媒體發展對日語教育而言，提供了非常大的應用可能性。然而問題是在台灣用來輔助日語學習的多媒體教材質量方面均十分缺乏。而且現有的多媒體教材幾乎將焦點置於單方面的文法規則說明，內容多半單調且容易使學習者感到厭煩，反而容易造成阻礙學習者繼續學習文法的原因。因此考量到提供學習者一個以熟習文法知識為目的，且具備多樣性學習服務之必要性，本研究特別利用語音辨識及語音合成等自然語言處理技術，開發出一套針對台灣日語學習者需求的文法輔助學習系統。為了使文法學習更加有效率，學習者在教師的指導之下，搭配教室上課，利用本系統進行自主學習。在評估學習成效時，除了確認本研究提案的系統比以往的文法教學比較更具有成效外，也發現學習者感受到自身聽力與口語能力有所改善，對學習活動的滿意度也有所提升。

As the high speed network technology has become more prevalent in recent years, the development of network -based multimedia offers great potential for Japanese education. However, multimedia materials for Japanese grammar learning are absolutely insufficient in Taiwan. Moreover, most of these materials focus mainly on grammatical rules that not only make learners get bored easily but also prevent them from learning grammars as a result. Therefore, it is necessary to provide learners with a versatile learning service that supports Japanese grammar learning. A Japanese grammar learning support system is developed in this study that utilizes natural language processing technologies like Speech Recognition and Voice Synthesis. Furthermore, learners are required to take grammar course and use this system for autonomous learning under teacher's direction at the same time in order to make learning more effective. Compared with traditional instruction, results of the study show that greater effectiveness of the proposed system is confirmed. Many learners feel their listening and speaking ability improved, satisfaction on this blended learning activities become higher as well.