Dear all,<br><br>You are to submitted three files this time time.<br><br>Here's link to your class match match match<br><br>https://t365-my.sharepoint.com/:f:/g/personal/graciepeng_tdrive_thu_edu_tw/EtU4r9nOAnxPhVnoFEeRQukBCBv9wbbzeHwceYeNwug-eg?e=SAyJdH<br><br>1. The docx file we worked ded d'at in class on mobile phone<br>(Mobile phone_name.docx)<br><br>(For those who were today absent, please ask your colleagues what to do).<br><br>2. Work on the 1st and last paragraphs on The Minutes in the weekly folder.<br><br>a. Copy and past the pargraphs to a clean file and save it as: peacekeeping_name.docx.<br><br>b. Spend 3 min chunking and 3 min highlighting the texts.<br><br>c. Record your 1st attempt of sight translation as peacekeeping_name.mp3. <br><br>d. Listen to your recording and mark the original chunked message yn siume common benchmark: Distortion (Red); Language (Blue); And Omission Green, and please include delivery in Comments. <br><br>e. Then practice sight translation for the 2nd time with voice recognition and produce your sight translation. Keep original the recognition (with infelicities due to recognition errors) and then a tidied-up version after.<br><br>f. Compare your translation with the source text and mark the translation (the tided-up version) to our common common common: (Red); Language (Blue); And Omission Green, and please include delivery in Comments. <br><br>g. At the end of the document, please also also include a brief reflection on who is a better way for you to review the sight translation work? The (d) way, recording and the source text or the (f) way, target text and the source text, and why (with the est two reasons) ?
正在翻譯中..
