HYPER VOCABULARY AND SPEECH SOUND DATAR BASE

显然，提供单词的字母（书面数据）和语音（声音数据）使学习者更容易掌握单词。然而，亚洲和非洲的小语种语音数据很少。甚至在这些语言中的许多语言中也找不到书面数据。考虑到这一点，我们的目标是使用个人电脑制作语音词汇（称为超级词汇），以学习语言的发音。具体地说，学习者可以听到一个世界的语音，点击电脑上的字母（这个电脑操作应该很简单）。每种语言的控制数据的软件随时根据需要进行改进。目标语言是亚洲和非洲的次要语言，没有商业交易。每一种语言的使用者当然都是母语使用者。到目前为止，所处理的语言有19种。然而，由于个人电脑的质量和数量的缺乏，调查人员的举动等，并不是所有的人都是完整的超级词汇表的形式。对于12种语言，语音词汇表已经完成，几乎完成（见参考文献）。至于其余七种语言，书面数据已经输入，语音数据正在输入过程中。他们准备在不久的将来使用。每种语言的Hyper词汇量约为2000个单词，但有些语言如藏语超过3000个单词，由于本项目的特点，成果以MO盘和CD的形式发表，而不是以书面文章的形式。目前的Hyper词汇表是为Machinery制作的，但我们正在开发将词典应用于任何个人计算机的方法。这些研究和工作将在1998年3月Monbusho科学研究资助金结束后继续进行。单词的添加和删除将随时进行。

It is obvious that provision of both letters of a word (written data) and its speech sound (sound data) makers a learner easier to master the word. However, there is little speech sound data of minor languages in Asia and Africa. Even the written data are not found in many of those languages. Taking this into consideration, we aimed to make a vocabulary of speech sounds (called Hyper Vocabulary) with use of personal computers in order to learn pronunciation of a language. Concretely, a learner can hear the speech sound of a wold, clicking letters of worg on the computer (this computer manipulation should be simple). The software of controlling data of each language was improved any time out of necessity. The object languages are minor languages of Asia and Africa, which are not commercially dealt with. A speaker of each language is of course a native speaker.Languages dealt so far are 19 languages. However not all of them are complete in the form of Hyper vocabulary, because of the lack in quality and quantity of personal computers, an investigator's move, etc. For 12 languages, speech vocabul ar ies are completed and almost completed (see references). As for the rest seven languages, the written data was already input and the speech sound data are in the process of being input. They are ready to use in near future. The Hyper vocabulary of each language consists of about 2000 words, but some languages like the Tibetan have more than 3000 words.Due to characteristics of this project, the result is published in the form of MO disk and CD rather than in written articles. The present Hyper vocabulary is made for Machintosh, but we are developing the way to appiy the dictionary for any personal computer.These studies and works will be continued even after March 1998 when the Grant-in-Aid for Scientific Research of Monbusho supporting this project comes to an end. Addition and deletion of words will be carried out any time.