Building a multi-media database for language research and education

在第二年的研究中，我们的研究集中在A.进一步积累字符数据，B.制作可操作的文本搜索程序，和C.积累对话数据。对于目标A，我们通过从CD-ROM中接收电子数据积累了大约1亿字。目标B的工作已经取得进展，但还没有完全完成。这样我们就能大致了解它会是什么样子我们的文本搜索程序的完成是需要进一步研究的领域之一。至于C语言，它一直是一个有问题的领域，因为带有封闭式字幕对话数据的激光盘被证明是很难获得的，而且因为译文不是逐字逐句的。总之，我们认为我们已经取得了相当大的进步，鉴于计算机技术的迅速发展，这往往不利于建立像我们这样的数据库系统。

In its second year of research, funded by this program, our studies focused on A.further accumulation of character data, B.production of workable text-seach program, and C.acumulation of dialogue data.As to the Aim A,we have amassed roughly 100 million words by taking in electronic data from CD-ROM's.Work has progressed but not quite finished with Aim B,so that we only have a rough idea of what it's going to be like. The completion of our text-search program is one of the areas for further study.As for C,it has been a problematic area because LaserDiscs with closed-caption dialogue data proved to be difficult to obtain and also because the transcriptions were not word for word.All in all, we think we have made a considerable progress, in view of the rapid progress in computer technology, which more often than not works against building of database system such as ours.