紀要論文

紀要論文

text

大阪大学高等教育研究. 2014, 2, p. 11-18

日本語

text

2014-03-31

In this paper, I report on an example of German voice training using the speech recognition application“Dragon Dictation” and provide results of research on how the German sentences Japanese college students pronounced are recognized by the application. A total of 114 students took part in the study (91 from Osaka University and 23 from Konan University). They learn German as a second foreign language. The training was held during the first 30 minutes of the class, and the students spoke sentences aloud to“Dragon Dictation”and wrote down the dictated text on a practice paper. In the first part of the research, one of six example sentences,“Am Dienstag um zwei treffe ich Freunde.”, was dictated almost always correctly by Dragon Dictation. In contrast, only a few students could pronounce the following sentences correctly: “Was haben Sie am Mittwoch um Viertel vor sechs vor?”“Haben Sie am Mittwoch von zwei bis vier Zeit?”, and a lot of them had the same pronunciation problems with“um Viertel”and“vier”. Other words of these sentences were dictated almost perfectly. An analysis of the practice papers revealed that a few learners successfully tried to change their German pronunciation without any instruction. In the next part of the research, which took place from October to November at Konan University, it became clear that there are some German sentences that Japanese learners could speak without any pronunciation problems, such as“Gehen wir morgen Abend ins Kino?”“Ich trinke Tee mit Milch.”and two sentences which they could barely pronounce correctly: "Ich suche eine Bluse.”“Mögen Sie Zwiebeln?”There are, however, some cases in which I cannot determine the checkpoint from the dictated texts. I have to research this issue further.

10.18910/28099

28099

音声認識アプリを用いたドイツ語発音学習の実践と検証

岩居, 弘樹

大阪大学全学教育推進機構

日本語

2014-03-31

大阪大学高等教育研究

2

11

18

https://hdl.handle.net/11094/28099

10.18910/28099

https://doi.org/10.18910/28099