语言的电子资料库.pptVIP

Electronic Database for Linguistic and Language Research 語言的電子資料庫 和 語言研究 一、機器語言可處理的文件快速增加 原因 1.電子出版 2.線上資料庫及紀錄 二、基本語言分析 ↓ 基本語言資料庫研究 三、機器語言可處理的文件來源 1.易取得與使用的大型資料庫 2.可解決讀者、學者、學生的基本需求 Dictionaries 字辭典 事實上，現代字典都是電子化生產，並且可以以機器呈現、輸出。二十年前，第一部可以機器閱讀的電子字典是1968年約翰歐尼的webster’s第七版字典。 電子字彙資料庫的發展可以回溯到那時期。 American Heritage Electronic Dictionary 美國遺產電子字典 Longman Dictionary of Contemporary 朗文電子字典 CD-ROM of the Oxford English Dictionary牛津唯讀電子字典 Languages of the World 語言世界 一種名為”語言世界”的字典可以辨認十八種世界上的語言，其系統資源使用者收尋或選擇跨語言的複合字詞。 Over fifty scholars from Europe, the Middle East, and North America are participating in the TEI efforts to define sets of tags for marking features of texts and to code the tag sets within the framework of the Standard Generalized Markup Language (SGML). SPEECH DATABASES 語音資料庫 Instruction Speech recognition and synthesis efforts in the 1980s plus advances in storage technology have led to establishment of shared speech databases that can be used for algorithm development and system evaluation and testing. Such databases also provide a common reference for evaluation across systems and techniques. Speech data differ considerably from text data. The DARPA-supported spoken language systems research has resulted in several speech databases, including the large TIMIT database developed by Texas Instruments and MIT. The JEIDA Japanese language database A similar database for French is GRECO, representing speech from thirty-two speakers. It is clear that the availability of speech databases is not as advanced as that of text databases because of the logistics and resources associated with speech recording under controlled conditions. ※電腦科技促進語言資料潛力的發展下所產生的議題 1. for研究人員使用的方法學 2. 標準化的架構 3. 促進交流和研究合作 4. 專業出版品的價值 第一組組員 495100138 吳敏薇 495100152 楊念穎