A7 4^83 7 9 -^-___B7_______ 五、發明說明（1) 螢明背景 1·發明範圍 本發明一般係關於外國語文教學，並且尤指以多媒體電 腦輔助系統來協助學生在模擬之會話環境中學習外國語 文。 2.背景 自古以來’學習外國語文可說是全世界人類所共同追求 的。時至今日這種追求比以往變得更加普遍與重要。當地 球村彼此之間愈來愈相關，國際旅行及商務愈來愈一般化 ’且世界透過通稱為全球資訊網（World Wide Web)而變得 愈來愈相互連接時，能夠與來自其他國度和文化的人交談 就成為一種必要性而非學術性的工作了，的確，由於英語 已成為國際商用語言’故英語學習乃為全世界已發展與發 展中國家無可避免的狀況。譬如說，在日本的中小學，不 僅僅必須要學習多年的英語課堂課程，並且另外增加私人 英語教學也是相當常見。 只有當出訪美國以外地區後’才會驚訝於其他國家的成 人與孩童’花在學習英語的時間以及努力的程度。即使在 美國，學習外國語文也是一個大家共同的目標。 到目前為止，學習外語的程序一直依照著一種或多種古 早的方式：冑堂上老師利用面對面互冑與課本教材的帝式 來教學，帶領著學生完成整個學習程序；自修用錄音帶、 錄影帶’可播放字囊及片語並且能夠暫停以便學生對放音 設備覆誦之° ' -4- ^纸奋尺度適用中國國家標準（CNS)A4規格（210 x 297公爱）' ------ C請先閱讀背面之ii意事項再填寫本頁) I ^ · n 1 n n n n-*-r-°J :: I - 經濟部智慧財產局員工消費合作社印製 “48379 A7 B7 五、發明說明（2 隨著個人電腦的進步，現在已可取用外語敎學軟體。美 國專利第5,8】0,599及5,697,789號所述即爲〆例。然而， 目前可使用的程式，基本上依舊只是以前常用類似像卡片 和帶子等非互動性媒介的轉型，偶爾加上一些娱樂性或敎 育性的圖形或視效（例如，第789號專利裏當杀範者發音時 ，會利用動畫影像效果來表現其唇部動作）。不過，當前所 能使用的外語敎學軟體，並不能提供眞正能夠近似於或模 擬教師-學生互動環境的互動式系統。 在所有目前能夠獲得的敎學系統中，一般公認敎師·學生 互動式敎學法（必須是良好訓練及經驗豐富的敎師），而且 最好是一對一的環境，才是最適合與最有效率的外語學習 方式，並且是遠勝於利用卡片或帶子Λ但是很不幸地，大 多數的敎學與互動方式都沒有辨法達到上述的優點。 面對面互動式敎學最爲合適乃基於下列數個理由。首先 ，在敎師-學生互動式的環境裏，才能有眞正即時性（ReaI Time)且專爲該學生的程度、優點與有待加強之處所背景的 叉談。其次，敎師才能接收到學生各種的回應，其中或包 括學生錯誤的發音與文法結構，而依此提出正確的回簽， 並矯正學生所犯的錯誤部份。第三點，如此將產生廣知的 肢體-語言（Physio-linguistic)現象，也就是—個人由視覺所 見到的線索，會影響到他同時由聲音所接收到的。這也稱 爲McGurk效應，特別是當一個人在孩提時代，學習著以 一些語音來搭配嘴部/臉部的定位及表現的情況。換句話説 ，當一個人將視覺所見到的線索與聲音同時合併，即可大 本紙張尺度適用令囡國家標準（CNS)A4規格（2_〗0 X 297公爱了 i <請先閲讀背面之注意事項再填寫本頁) . 經濟部智慧財產局員工消費合作社印製 4483^9 經濟部智慧財產局員工消費合作社印製 A7 B7 五、發明説明（3) 幅增加通曉及明瞭的能力。商業人士多偏好面對面會議， 而不是電話商談，即可明顯看出這種現象。雖然後者比較 方便，但是大多數仍舊相信面對面會議的溝通，一般來説 比電話商談爲佳，因爲就算每一個所説字都聽得非常清楚 ，電話商談裏還是缺少了視覺的線索。 不過，面對面的敎師-學生系統也是有它的缺點。因爲在 敎室環境内有許多學生，所以不能進行眞正的個人化敎學 法。此外’在敎室的環境裏，意味著要將許多人同一時間 集合到同一個地點。這表示需要排程，边使學生調整個人 的時間以配合課表，而不是只要一通電話老師就來。而敎 室系統普通一周也只能上最多幾個小時的課。因此，該系 統無法眞正地提供個人化的敎育經驗。 雖然家敎的方式可彌補部分的缺點，但是卻無法完全地 解決’而同時卻又還會產生自己的問題。舉例來説，即使 家敎系統只會牵涉到兩個人，卻也會有排定時程的問題。 而且價格較昴貴，敎師也不一定在任何時間或場合都隨時 可由學生呼叫到。 因此之故，本行業即有將上述各種優點集中，而又將缺 點降至最低，以藉此改善外語敎學系統的必要。 發明概述 本項經改良之系統，係以使用可購得並具有作爲通稱爲 多媒體計算能力的個人電腦.這種方式，來提供改善的方法 ’其中該方法包括了幾個步驟，諸如儲存到電腦的記憶體（ 或可存取到的位置）以便在螢幕上顯示眞實生活中背景類 * 6 - 本紙張尺料用中關家標準（CNS)A4规格（2】Qx 297公爱) ------ ^1 ♦-1 n n tt «^1 I. n >^1 n 』^I n I— 1 (請先間讀背面之注意事項再填寫本頁) 448379 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明（4 ) 似像速食餐廳、銀行、醫生診療室、等等的視覺模擬；儲 存到電腦的記憶體（或可存取到的位置）以便在發幕上顯示 一適合眞實生活的動畫人物，就像是上述速食餐廳裏的接 單服務員、銀行裏的櫃台行員、接待員或是診療室裏的護 士、醫生等；該改良系統將適合該顯示出來場所的主角之 會話敘述句儲存在記憶體裏；該改良系統將儲存到電腦的 記憶體（或可存取到的位置）以便透過電腦的揚聲系統來播 放學生對主角每一個會話敘述所可能提出的回答集 (Library);該改良系統由主角發起與學生間的外語會話， 且該會話適合一顯示出的背景値；該改良系統讓學以外ΐ吾 來回答主角；該改良系統利用儲存於電腦記憶體内（或可存 取到的位置）的自動語音辨識（Automatic Speech Recognition)技術，來將學生的口頭回答轉化爲文字；該改 良系統利用儲存於電腦記憶體内（或可存取到的位置）的自 然語言通曉（NaturalLanguage Understanding)技術，將轉化 過的回簽與存放於電腦記憶體内（或可存取到的位置）學生 可能答案的資料館進行比對，以便將該回答的意義解譯出 來；該改良系統由該資料館選出對主角最適當的回應：該 改良系統啓動該應答，將之透過電腦的揚聲系統來播放， 並且與主角的「嘴部」視效同步；以本方式繼績輪流進行 多個會話，包括主角的敎述或詢問，再由學生回答或是主 角回答，如此直到整個適合於該背景的會話完畢。 藉由以本創新、非顯而易見的多種技術合併應用方式， 透過多媒體電腦輔助系統，可提供外語敎學法一模出近乎A7 4 ^ 83 7 9-^ -___ B7_______ 5. Description of the invention (1) Fluorescent background 1. Scope of the invention The present invention is generally related to the teaching of foreign languages, and especially multimedia computer assisted systems to assist students in a simulated conversation environment Learn foreign languages. 2. Background Since ancient times, 'learning foreign languages' can be said to have been pursued by humans all over the world. This pursuit is more common and important than ever before. As global villages become more and more related to each other, international travel and commerce are becoming more generalized, and the world is becoming more and more interconnected through what is commonly known as the World Wide Web, it is able to connect with other countries and cultures. Talking with people has become a necessity rather than an academic job. Indeed, because English has become an international business language, English learning is an unavoidable condition for the world's developed and developing countries. For example, in Japanese elementary and junior high schools, it is not only necessary to study English classes for many years, but it is also quite common to increase private English teaching. It ’s only after visiting the United States that I ’m surprised that adults and children in other countries ’time and effort spent learning English. Even in the United States, learning a foreign language is a common goal. So far, the process of learning foreign languages has been in accordance with one or more ancient ways: teachers in the classroom use imperial style of face-to-face interaction with textbooks to lead students through the entire learning process; self-study audio and video Can play the word bag and phrase and can be paused for students to recite the playback device ° '-4- ^ Paper Fen standard applies Chinese National Standard (CNS) A4 specification (210 x 297 public love)' ----- -C Please read the meanings on the back before filling this page) I ^ · n 1 nnn n-*-r- ° J :: I-Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs "48379 A7 B7 V. Invention Explanation (2 With the advancement of personal computers, foreign language learning software is now available. U.S. Patent Nos. 5,8] 0,599 and 5,697,789 are examples. However, the currently available programs are basically still only the ones used to be The transformation of non-interactive media such as cards and tapes is often used, and occasionally some entertainment or educational graphics or visual effects are added (for example, when the killer pronounces in Patent No. 789, animation image effects are used to Express their lips Action). However, currently available foreign language learning software does not provide interactive systems that can approximate or simulate the teacher-student interaction environment. Of all the currently available learning systems, generally recognized teachers · Students' interactive learning method (must be a well-trained and experienced teacher), and preferably a one-to-one environment is the most suitable and efficient foreign language learning method, and is far better than using cards Or belt Λ But unfortunately, most of the philosophies and interaction methods do not achieve the above-mentioned advantages. Face-to-face interactive philology is most suitable for the following reasons. First, in the teacher-student interactive Only in this environment can ReaI Time be specific to the student ’s level, merit, and background to be strengthened. Second, the teacher can receive a variety of student responses, including student errors Pronunciation and grammatical structure, and based on this, propose correct check-backs and correct the mistakes made by students. Third, this will produce the well-known body-language (P (hysio-linguistic) phenomenon, that is, the cues that an individual sees through vision will affect what he receives at the same time by sound. This is also called the McGurk effect, especially when a person learns to use some speech in childhood Match the positioning and performance of the mouth / face. In other words, when a person combines visual cues and sounds at the same time, the paper size can be applied to the national standard (CNS) A4 specification (2_ 〖0 X 297 public love i < please read the notes on the back before filling out this page). Printed by the Employees 'Cooperatives of the Intellectual Property Bureau of the Ministry of Economy 4483 ^ 9 Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economy A7 B7 (3) Increase the ability to understand and understand. Business people are more inclined to meet face-to-face than to talk over the phone. Although the latter is more convenient, most people still believe in face-to-face communication. Generally speaking, it is better than talking over the phone, because even if every word spoken is very clear, there is still no visual clue in the talking. However, the face-to-face teacher-student system also has its disadvantages. Because there are many students in the classroom environment, it is not possible to conduct a personalised learning method. In addition, in the context of a cell, it means bringing many people together at the same place at the same time. This means scheduling is required while students adjust their personal time to match the schedule, instead of just having to call the teacher. In general, the room system can only take less than a few hours a week. As a result, the system cannot properly provide a personalized education experience. Although the furniture method can make up for some of the shortcomings, it can't completely solve it, and at the same time, it will also create its own problems. For example, even if the furniture system involves only two people, there will still be scheduling issues. Moreover, the price is relatively expensive, and the teacher may not always be called by the students at any time or occasion. Therefore, it is necessary for the industry to concentrate the above-mentioned various advantages while minimizing the shortcomings in order to improve the foreign language learning system. SUMMARY OF THE INVENTION This improved system is based on the use of a personal computer that is commercially available and has what is commonly referred to as multimedia computing power. In this way, an improved method is provided, wherein the method includes several steps, such as storing to a computer Memory (or accessible location) to display real-life backgrounds on the screen * 6-This paper size uses Zhongguanjia Standard (CNS) A4 specifications (2) Qx 297 public love) --- --- ^ 1 ♦ -1 nn tt «^ 1 I. n > ^ 1 n』 ^ I n I— 1 (Please read the precautions on the back before filling this page) 448379 A7 B7 Intellectual Property Bureau, Ministry of Economic Affairs Printed by the employee consumer cooperative V. Description of the invention (4) Visual simulations like fast-food restaurants, banks, doctors' clinics, etc .; stored in the computer's memory (or accessible location) for the launch Display an animated character suitable for real life, like the order server in the fast-food restaurant, the counter staff in the bank, the receptionist, or the nurse or doctor in the consultation room; the improved system will be suitable for the displayed place The protagonist's conversation narrative is stored in Memory; the improved system will be stored in the computer's memory (or accessible location) in order to play through the computer's speaker system, the student's answer set (Library) that each protagonist may give to each conversation; The improved system is initiated by the protagonist in a foreign language conversation with students, and the conversation is suitable for a displayed background; the improved system allows non-school students to answer the protagonist; the improved system uses storage in computer memory (or accessible) To the location) 's Automatic Speech Recognition technology to convert students' spoken answers into text; the improved system uses natural language knowledge (NaturalLanguage) stored in computer memory (or accessible location) Understanding) technology, which compares the converted check-back with a library of possible answers stored in the computer's memory (or accessible location) to interpret the meaning of the answer; the improved system consists of The library selected the most appropriate response to the protagonist: the improved system activates the response and sends it through the computer's speaker system Release, and synchronize with the protagonist's "mouth" visual effects; in this way, multiple conversations are performed in turn, including the narration or inquiry of the protagonist, and then answered by the student or the protagonist, so that until the whole The session is complete. By merging applications with multiple innovative and non-obvious technologies, through multimedia computer-aided systems, it can provide foreign language learning methods almost exactly
本紙張尺度適用中國國家標準（CNS)A4規袼（210 X 297公釐） -------i ---—I— ----!1*—訂 - - - ------I , (請先閱讀背面之注意事項再填寫本頁) 4 483 7 9 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明（5 ) 興實的敎師-學生互動環境，而該電腦輔助系統可在學生自 己的電腦上’同時地享受眞正的互動學習環境’以及隨時 隨地與時間長短自訂的個人便利。本發明可提供學生與各 種不同的「指導老師」之間模擬性、情境化的交談：連接 到伺服器的電話；獨立的電腦或其他裝置；或是連接到網 際網路的個人電腦。 舉例來説，某擁有一標準個人電腦配備的學生，即可「 輸入」速食餐廳並且與螢幕上的主角進行實況會話。當主 角開場對學生表示歡迎之意且提出第一個問題後（譬如説 『您好’歡迎光臨Cyber咖屋。有我任何能夠爲您服務 之處嗎？J )，並且系统將等待學生的口語回答。藉由自動 語音辨識及自然語言通曉技術’學生的回答將被電腦「解 譯J，並和預先設計之可能答案或/與知識表示法基底的集 合相互比對。相對以往傳統語言敎學系統要求確切的答覆 方式，本發明可接受在該主題範圍内符合文法（或反應）的 各種答案◊的確’本發明能夠接受數百種甚至上千種可能 的語音回答輸入’無論是正確或不正確，然後系統根據對 學生所回答的瞭解再回應給學生。這些回應可包括一答案 、要求提供更進一步的資料、予以更正、拒絕、提示等等 〇 透過電腦的揚聲系統，可與學生以語音「交談」，並且 螢幕上的動畫主角，其臉部表情會與喇u八的聲音輸出同步 ，也可以提供給學生適當的視覺線索模擬。本發明也包括 了 一會話「樹（Tree)」’可使學生與主角之間的互動得以依 -8- 本紙張尺度滷用中國國家標準（CNS)A4規格（210 * 297公釐）This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) ------- i ----- I-- ----! 1 * —Order------- --I, (Please read the precautions on the back before filling this page) 4 483 7 9 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. 5. Description of the invention (5) The practical teacher-student interaction environment, The computer-assisted system can 'enjoy a positive interactive learning environment at the same time' on the student's own computer and personal convenience that can be customized anytime, anywhere, and length of time. The invention can provide simulated and contextual conversations between students and various "mentors": a phone connected to a server; a stand-alone computer or other device; or a personal computer connected to the Internet. For example, a student with a standard personal computer can "enter" a fast food restaurant and have a live conversation with the protagonist on the screen. After the protagonist welcomes the students and asks the first question (such as "Hello", welcome to Cyber Cafe. Is there anything I can do for you? J), and the system will wait for the students to speak Reply. With automatic speech recognition and natural language proficiency technology, the student's answer will be interpreted by the computer "J, and compared with a pre-designed set of possible answers or / and a set of knowledge representation bases. Compared to previous traditional language learning system requirements Exact way of answering, the present invention can accept a variety of grammar (or response) answers within the scope of this subject. Indeed, the present invention can accept hundreds or even thousands of possible voice answer inputs, whether correct or incorrect, and then The system responds to the students based on their understanding of the answers. These responses may include an answer, request for further information, corrections, rejections, prompts, etc. 0 through the computer's speaker system, you can talk to the students by voice And the animated protagonist on the screen, the facial expression will be synchronized with the voice output of La Ba, and it can also provide students with appropriate visual clue simulation. The present invention also includes a conversation "Tree" which enables students to interact with the protagonist in accordance with the Chinese National Standard (CNS) A4 specification (210 * 297 mm).
In I ·1» n n n I * VI ·1 Hr n 一.eJ> n It n I 線丨 <請先閱讀背面之注意事項再填寫本頁) ίϊ 43379 Α7 Β7 五、發明說明（6 ) 循一專爲該學生而設計之路控。例如説，在Cyber咖啡屋 裏’如果學生想點一客漢堡，則學生可以有不同的説法_ 譬如「我想要一份漢堡」，「請給我一客漢堡」，或是「 我要一個起士漢堡」。這些都會被電腦認定爲適當的回答 ’且電腦對比還會詢問學生肉片想要幾分熟，是否要加洋 葱及酸黃瓜’要不要薯條等等。不過，如果學生點的是披 薩’那麼本發明的方法式會辨識該回答，並詢問要厚片抑 或薄片、要加什麼料的披薩。 較佳實施例之詳細説明 在本較佳實施例中，本發明之方法係使用一般可購得的 多媒體個人電腦’該電腦具有處理單元、記憶體、監視器 、作業系統、CD-ROM光碟機、鍵盤、滑鼠（或其他等似裝 置）'揚聲器及麥克風等，如同一般由IBM、Compaq、Dell 與Gateway等各廠商廣泛銷售之電腦設備。 電腦記憶體係以傳統方式先行載入自動語音辨識（ASR) 及自然語言通曉（NLU)引擎。各種不同的ASR技術已商業 化並可購得。這些系统裏，有一些是與揚聲器相關，意思 就是彼等一直以來係依辨識由單一喇P八的造句及發音而製 。某些系统會辨識離散言談模型，其中會以短暫中斷而將 述句分隔開來。另外有些系統則是能夠辨識連續性的言談 。此外’有些系统擁有小型的字彙與資料庫，有些則有相 當大的字彙與片語資料庫。不過，無論是何種系統，都能 夠將口述的文句加以轉化，並分析該文句以決定其意義。 在本較佳實施例中，使用Unisys公司所提供的ASR及 -9- 本紙張尺度適用中园國家標準（CNS)A4規格（210 X 297公蹵） ______________--- <請先閱讀背面之ii意事項再填寫本頁)In I · 1 »nnn I * VI · 1 Hr n a. EJ > n It n I line 丨 < Please read the precautions on the back before filling in this page) ίϊ 43379 Α7 Β7 V. Description of the invention (6) Follow Road control designed for this student. For example, in the Cyber Cafe, 'If a student wants to order a hamburger, the student can say different things_ such as "I want a burger," "Please give me a burger," or "I want a burger." Cheeseburger. " These will be regarded as appropriate answers by the computer, and the computer comparison will also ask students how much meat they want to cook, whether to add shallots and pickles, or fries, etc. However, if the student orders pizza, then the method of the present invention recognizes the answer and asks for thick or thin slices and what kind of pizza to add. Detailed description of the preferred embodiment In the preferred embodiment, the method of the present invention uses a generally available multimedia personal computer. The computer has a processing unit, a memory, a monitor, an operating system, and a CD-ROM drive. , Keyboard, mouse (or other similar device) 'speakers and microphones, like computer equipment generally sold by various manufacturers such as IBM, Compaq, Dell and Gateway. The computer memory system traditionally loads the automatic speech recognition (ASR) and natural language literacy (NLU) engines first. Various ASR technologies have been commercialized and are commercially available. Some of these systems are related to speakers, which means that they have always been made by recognizing the sentence formation and pronunciation of a single La P eight. Some systems recognize discrete speech models in which sentences are separated by short interruptions. Other systems are capable of recognizing continuous speech. In addition, some systems have a small vocabulary and database, while others have a relatively large vocabulary and phrase database. However, no matter what the system is, it is enough to transform the spoken sentence and analyze the sentence to determine its meaning. In the preferred embodiment, the ASR and -9 provided by Unisys are used. This paper size is applicable to the China National Standard (CNS) A4 specification (210 X 297 cm). ______________ --- Please read the back (ii) Please fill in this page if you have any questions.
-I-6J -線· 經濟部智慧財產局員工消費合作社印製 448379 A7 -------B7五、發明說明（7 ) 經濟部智慧財產局員工消費合作社印製 NLU產品’搭配 Unisys Natural Language Assistant (NLA) 使用，並且按照製造廠商的規格及指示儲存在電腦記憶體 内。這些可購的技術產品，使得電腦能夠辨識及解譯譬如 説英語會話，可以讓互動性自然地產生，而不會有複雜的 選單或是做作的字彙語法限制。 電腦記憶體也由一般方式載入程式，以便在監视器上顯 π眞實生活裏，某人出旅所習語言之國家時，常會遇到的 情境背景視覺模擬。舉例來説，如果本發明方法係用於敎 授英語時，情境背景會包括有速食餐廳 '銀行、機場、郵 局、診所、汽車出租公司、旅館或公車等。而熟悉該技藝 工人士應明瞭，旅客將會遇到極多的情境背景，而在情境 背景中會有其相關之交談用字彙。藉由學習者自己的電腦 ，學習者在眞正步入美國的一家速食餐廳之前，就早已多 次上過虚擬的速食餐應並且順利點餐，這就將會使得學習 者比起那些只有背誦過英文漢堡、薯條和可樂單字的旅 ，更能處理眞實生活情況。 電腦記憶體也以傳統方式存放有關資料，以便於監視器 該環境裏顯示出適合該環境的「主角」。例如，假設背景 爲速食餐靡，主角會是接單員，站在櫃檯與結帳處後方工 作，並會向顧客（在此爲學生本人）致意歡迎，接受訂單， 取得並遞交食物予學生’説明總價，收費並找回零錢。如 背景爲銀行’那麼主角就是行員，同時也會向顧客即學生 本人表示歡迎’詢問並且接受學生的輸入而進行所 易’然後完成該筆交易作業。 (請先閱讀背面之注意事項再填寫本頁) ϋ.Λ . i線· _*10" 本纸張尺度適用中國國家標準（CNS)A4規格（2〗〇 X 297公爱） 448379 經濟部智慧財產肩員工消費合作社印製 A7 B7 五、發明說明（8 )-I-6J-Line · Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 448379 A7 ------- B7 V. Invention Description (7) NLU products printed by the Employees’ Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs' coordinated with Unisys Natural Language Assistant (NLA) is used and stored in computer memory in accordance with the manufacturer's specifications and instructions. These commercially available technology products allow computers to recognize and interpret, for example, English conversations, allowing interaction to occur naturally without the need for complicated menus or constrained word grammar. Computer memory is also loaded into the program in a general way to display on the monitor the visual simulation of the contextual background often encountered when someone travels to a country where he is studying a language. For example, if the method of the present invention is used to teach English, the context may include a fast food restaurant 'bank, airport, post office, clinic, car rental company, hotel, or bus. Those skilled in the art should understand that passengers will encounter a lot of contextual backgrounds, and there will be related words in the contextual context. With the learner's own computer, the learner had already served virtual fast food meals and ordered successfully many times before he entered a fast food restaurant in the United States, which will make the learner more than those Only trips that recite the words hamburger, fries, and cola in English can handle real life situations better. Computer memory also stores related data in a traditional way so that the monitor can display the "protagonist" suitable for the environment. For example, suppose the background is fast food, and the protagonist will be the order taker, standing behind the counter and the checkout office, and will greet the customer (here is the student himself), accept the order, obtain and submit the food to the student 'State the total price, charge it and get the change back. If the background is a bank, then the protagonist is a clerk. At the same time, the customer, the student himself, will be welcomed to inquire and accept the student's input to make the transaction, and then complete the transaction. (Please read the precautions on the back before filling in this page) ϋ.Λ. I-line · _ * 10 " This paper size applies to China National Standard (CNS) A4 (2〗 〇〇297 爱爱) 448379 Ministry of Economic Affairs Wisdom A7 B7 printed by the property consumer employee cooperative Fifth, the invention description (8)
X 對於每一個虛擬背景，電腦記憶體以一般方式存有一適 合於該背景的交談樹（Conversational Tree)。該交談樹係由 文字與片語字彙所組成，能夠模擬主角及學生之間的會話 爲符合該背景的眞實生活情況。該會話將由主角發出。例 如’在Cyber咖啡屋中，主角會先歡迎學生，然後詢問要 點些什麼。藉由ASR、NLU及NLA技術，電腦可以辨識 出許多不同的答覆。如果學生的回答被辨識爲任何有關於 點選漢堡的敘述，那麼就會追蹤到交談樹關於點選漢堡的 「枝幹」上去。爲回應學生點選漢堡的敘述，主角還可以 被設計成再詢問學生想要什麼樣的漢堡。一個適當的回答（ 同樣藉由ASR、NLU及NL A技術），引發主角下一個合乎 邏輯且適當的敘述或是問句—譬如説詢問學生要不要加洋 蔥。 如果電腦無法瞭解學生第一個答覆或是回答有誤，則電 腦會被設計爲重覆詢問以再次獲得答覆。經過一定的誤失 或非正確答覆預設次數’電腦可以學生的母語來翻譯主角 的叙述或是問句，然後再重新開始會話。 在會話中，每一種語言都有其對於溝通極爲重要的視覺 線索，話語及適當的視覺線索之組合，可以提供非常有效 的溝通。爲強化在本發明方法之較佳實施例中的學習經驗 ，電腦記憶體也以傳統的方法存有資料，使得主角同時能 夠與電腦所發出的聲音「對嘴」，以提供一般在該敍述裏 ，所伴隨而來的視覺線索。一個簡單的例子，在速食餐廳 會話的某個過程當中’主角可能會對顧客（學生）説明紙巾' _____- 11 - 本紙張-------------------------------- :___.1(-----1、」".1 I (請先閱讀背面之注意事項再填寫本頁} .. -線. 4S373 A7 B7 五、發明說明（9 ) 的位置：「就在那裏」。眞實生活裏，服務生很可能會用 手來指到正確的方向。在本發明方法之較佳實施例中，模 擬的主角當説明時也會同樣地用手指著。 本發明的方法會包括任何數目的背景，該背景可由電腦 磁碟或CD-ROM來存取。學生可體驗到外國的模擬旅行。 例如，使用本發明的方法，一位日本商務人士可以虛擬的 方式出訪美國，並且眞正地踏上美國之前，就已經在機場 、旅館、餐廳、銀行或是商業會議上順利地溝通許多次了 〇 雖然本發明方法之較佳實施例如同上述，惟很明顯地熟 知該技術之人士應可修改本較佳實施例，而不會悖離前表 之本發明概念。因此’本專利與其所提供之保護應不侷限 於上述之具體實施例’而是下列之申請專利範圍並包括與 其相仿應用之範疇。 --I I ---—----訂--------- (請先閲讀背面之注意事項再填寫本頁) 經濟部智慧財產局員工消費合作社印製 本紙張尺度適用令國國家標準（CNS)A4規格（210 X 297公釐）X For each virtual background, computer memory generally stores a Conversational Tree suitable for that background. The conversation tree is composed of words and phrases, which can simulate the conversation between the protagonist and the students to fit the real life situation of this background. The conversation will be sent by the protagonist. For example, 'In the Cyber Cafe, the protagonist will welcome the student first and then ask what to order. With ASR, NLU and NLA technology, the computer can recognize many different responses. If the student's response is identified as any narrative about clicking on a hamburger, it will be traced to the "branch" of the chat tree about clicking on a burger. In response to the student's narrative of choosing a burger, the protagonist can also be designed to ask the student what kind of burger he wants. An appropriate answer (also via ASR, NLU, and NLA technology) triggers the protagonist's next logical and appropriate narrative or question—for example, asking students if they want to add onions. If the computer cannot understand the student's first response or the answer is incorrect, the computer will be designed to repeat the query to get the response again. After a certain number of missed or incorrect responses, the computer can translate the protagonist's narration or question in the student's native language, and then restart the conversation. In conversation, each language has its visual cues that are extremely important for communication. The combination of discourse and appropriate visual cues can provide very effective communication. In order to strengthen the learning experience in the preferred embodiment of the method of the present invention, the computer memory also stores information in a traditional way, so that the protagonist can "talk" to the sound made by the computer at the same time, so as to provide the general description in the narrative. , The accompanying visual clues. A simple example, during a fast food restaurant conversation, 'the protagonist may explain paper towels to customers (students)' _____- 11-this paper --- ---------------: ___. 1 (----- 1, "" .1 I (Please read the precautions on the back before filling this page} .. -line 4S373 A7 B7 V. Location of the invention description (9): "There it is." In real life, the waiter is likely to use his hand to point to the correct direction. In a preferred embodiment of the method of the present invention, simulation The protagonist will also point his finger when explaining. The method of the present invention will include any number of backgrounds that can be accessed by a computer diskette or CD-ROM. Students can experience a simulated trip abroad. For example, using According to the method of the present invention, a Japanese business person can visit the United States in a virtual way, and before setting foot in the United States, he has successfully communicated many times at the airport, hotel, restaurant, bank or business meeting. The preferred embodiment of the inventive method is as described above, but it is obvious that those skilled in the art should be able to modify this preferred embodiment. Examples without departing from the concept of the invention described in the previous table. Therefore, 'this patent and the protection it provides should not be limited to the specific embodiments described above', but the scope of the following patent applications and includes the scope of similar applications. -II ----------- Order --------- (Please read the notes on the back before filling out this page) Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs Standard (CNS) A4 specification (210 X 297 mm)