TW527555B - Intelligent phonetic input system and method therefor - Google Patents

Intelligent phonetic input system and method therefor Download PDF

Info

Publication number
TW527555B
TW527555B TW88122058A TW88122058A TW527555B TW 527555 B TW527555 B TW 527555B TW 88122058 A TW88122058 A TW 88122058A TW 88122058 A TW88122058 A TW 88122058A TW 527555 B TW527555 B TW 527555B
Authority
TW
Taiwan
Prior art keywords
word
pronunciation
code
group
phonetic
Prior art date
Application number
TW88122058A
Other languages
Chinese (zh)
Inventor
Chiuan-Cheng Chiou
Hai-Biao Fan
Yun Wang
Original Assignee
Inventec Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inventec Corp filed Critical Inventec Corp
Priority to TW88122058A priority Critical patent/TW527555B/en
Application granted granted Critical
Publication of TW527555B publication Critical patent/TW527555B/en

Links

Landscapes

  • Document Processing Apparatus (AREA)

Abstract

An intelligent phonetic input system and method there for is disclosed here. First of all, classify the phonetic sign inputted by user into multiple phonemes. Subsequently follow multiple vowel combinations of phonemes in search of database to select word message with identical vowel combination, among them, each set of word message is comprised by all Chinese characters corresponding to the vowels. Then compare the phoneme of each set of word message with the inputted multiple phonemes to see if they are identical. When they are the same, display the word message for user to select.

Description

527555 A7 B7 五、發明說明() 發明領域: 本發明與一種注音輸入系統及其相關方法有關,特 (請先閱讀背面之注意事項再填冩本頁) 4¾别是一種可藉著輸入聲母而進行自動選擇字詞之智能注 太每 &只音輸入系統及方法。 --1 \ * 發明背景: 、厂 隨著電子科技日新月異的進步,由於各行各業皆朝 穴電子化、自動化努力,再加上近來網路使用之推波助瀾,527555 A7 B7 V. Description of the invention () Field of the invention: The present invention is related to a phonetic input system and related methods, especially (please read the notes on the back before filling this page) 4¾ It is not a type that can be input by initials. Intelligent note input system and method for automatically selecting words. --1 \ * Background of the Invention: With the rapid development of electronic technology, due to the electronic and automated efforts of all industries and industries, coupled with the recent use of the Internet,

►τ ^ I ^ V? 導致個人電腦大量普及。然而,對大多數的人而言,透過 、· ’之鍵盤輸入文字資訊至電腦中,依舊充滿了阻礙與困難。 一般而言,由於國語注音法則,爲國人所熟悉之中 文拼音方式,是以在對電腦或相關電子處理系統進行字詞 輸入時,注音輸入法爲I:常用的輸入方式。例如,在輸入 “中國”兩個漢字時,可藉著在键盤上键入“ i X人 « X汪” ,或是欲輸入“電腦”時,可在鍵盤上鍵入“为 一弓、)幺〆”,然後可由電腦螢幕上所出現具有上述讀 音的字串,選擇正確之漢字。 經濟部智慧財產局員工消費合作社印製 然而,在使用注音輸入法來進行輸入程序,仍然具 有許多缺點。其中,在國語注音法則裏,一個漢字的讀音 是由聲母、韻母與聲調所組成,是以在進行輸入程序時, 往往要在鍵盤上按動數次,才可鍵出所需之漢字。並且, 由於同樣的讀音,往往可代表數個不同的漢字,是以在键 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) 527555 A7 B7►τ ^ I ^ V? Leads to the widespread use of personal computers. However, for most people, entering text information into a computer through the keyboard of. Is still full of obstacles and difficulties. Generally speaking, due to the Chinese phonetic notation rule, Chinese Pinyin is familiar to Chinese people. When inputting words into a computer or related electronic processing system, the phonetic input method is I: a commonly used input method. For example, when entering the two Chinese characters "China", you can type "i Xren« X 汪 "on the keyboard, or you can type" for a bow, "on the keyboard when you want to enter" computer ". "〆", and then the correct Chinese character can be selected from the string with the above pronunciation appearing on the computer screen. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs However, there are still many shortcomings in using Zhuyin input method for input procedures. Among them, in the Chinese phonetic notation rule, the pronunciation of a Chinese character is composed of the initials, finals, and tones. When inputting a program, it is often necessary to press the keyboard several times to key out the required Chinese characters. And, because the same pronunciation can often represent several different Chinese characters, the Chinese paper standard (CNS) A4 (210 X 297 mm) is applied to the key paper size. 527555 A7 B7

修煩 正請 夺委 f Μ 系.: W ί:,平 :;(I 予3 杂所 正提 0之 五、發明說明( 入注音符號後,需再進行選字,以獲得所需之字詞。如此 一來,也使得使用注音輸入法來進行輸入程序之速度較 慢。 另外,由於中文注音輸入法,需要操作者輸入正確 的國語注音符號,是以當所欲輸入字詞之讀音,較不易掌 握時,往往會造成輸入錯誤。例如捲舌音“力與EJ ”、“ i 與下”和“尸與厶”,韻母“汪與亡”、“巧與尤”與“ ~與厶”, 或是中文聲調一聲、二聲、三聲、四聲、與輕聲等等,皆 容易爲操作者所混淆,並由此導致辨識錯誤而需重新輸 入。是以,對於不是非常熟悉國語注音法則之操作者而 言,會覺得使用傳統的中文注音輸入法並不方便。 發明目的及概述: (請先閱讀背面之注意事項再填寫本頁) 裝 文 中 之 母 韻 調 聲 略 省 可 種一 俣 提 在 的 目 之 明 發 本 法 入 輸 音 注 智 \lCul 種 1 供 提 明 發 本 時 序 程 入 輸 文 中 般一 行 進 在 中音 其注 。 入 統輸 系盤 及鍵 法由 方經 入先 輸首 音, 注 ,入 著輸 接組 0 一 音每 讀中 個其 數,、 爲碼 號編 符音 , 士買 音 、i 注入 的輸 入個 輸數 分爲 區音 可讀 ,個 著數 接述 。上 號換 符轉讀 · 經濟部智慧財產局員工消費合作社印製 母 聲 ,之 碼碼 編編 音音 讀讀 位入 進輸 二組 爲一 均每 碼 據 編根 音可 於 應 對 且 ’ 其 後。 然序 。程 音 讀 個 索 檢 行 進 之 中同 程 索 檢 此 自心 訊 .詞 字 相合 目符 數相 母母 聲聲 其其 碼碼 編編 音音 讀讀 入入 輸輸 與與 出出 分分 劃劃 先著 可接 序 本紙張尺度適用中國國家標準(CNS)A4規格(210 x 297公t ) 527555 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明( 修煩 正讀 太主有I 無扔 欠.7厂 予曰秦所 正提 〇之 之字詞訊息;再自儲存字詞訊息之位址,讀取字詞訊。隨 後,可決定每一組輸入讀音編碼之簡化條件,以確認輸入 讀音編碼其聲母、韻母與聲調之簡化狀況。再根據簡化條 件,對字詞訊息之每一組檢索讀音編碼進行簡化。接著, 比較字詞訊息之檢索讀音編碼與輸入讀音編碼是否相符 合。當相符時,則顯示於螢幕上,以供操作者選擇。 本發明並提供了一種用來輸入片語、短句之輸入方 法。其中,如同上述先輸入注音符號,並區分注音符號爲 數組讀音,再根據數組讀音中每一組讀音之聲母,進行檢 索程序,以自資料庫中選擇涵蓋上述聲母組合之字詞訊 息,其中每一組字詞訊息均涵蓋聲母組合。並且,比較字 詞訊息之讀音是否涵蓋數組輸入讀音。然後,將涵蓋輸入 讀音之所有字詞訊息,依照字數多寡顯示於螢幕上,以供 操作者選擇。 圖式簡單説明 : 藉由以下詳細之描述結合所附圖示,將可輕易的了 解上述内容及此項發明之諸多優點,其中·· 第一圖顯示根據本發明所提供對漢字讀音進行1 6 位元二進位編碼與儲存之方式; 第二圖顯示根據本發明所提供用以進行智能注音 輸入法之資料庫系統; 第三圖顯示根據本發明進行智能注音輸入法之主 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) ----------------- (請先閱讀背面之沒意事項再填寫本頁) · 527555 A7 五、發明說明() 要流*程圖; 第四圖顯示根據本發明進行智 旎注音輸入祛 聲每用實例; “爛八法之應 ft π 一 m岡觸不很诼尽發明進行知处 ¢.1時,於螢幕上所顯示之結果; 口此庄音輸 明 第五A〜B圖顯示根據本發 卜螢幕上所顯示之結果; 弟/、圖頌示根據本發明進行 冥 關流程; 短句或片 入法 語輸入之相 輸入之相 .第七圖顯示根據本發明進行 惹月關應用實例; ^句或片語 J 第八圖顯示根據本發明進行智〜、、 #建片語之相關流程;. --音輸入法來自 •ih徙 。之 第九A〜D圖顯示根據本發明淮广 以自建片語時,於螢幕上所顯示之红3此/王音輸入法 〜〜果;及 時 c請先閱讀背面之注意事項再填寫本頁) 第十圖顯示根據本發明所提供知 <配置圖。 曰 輸入系 统 • n ϋ n i n . 查^明詳細説明: 提 明 發本 經濟部智慧財產局員工消費合作社印製 著以有 藉可。 〇 ,的 統據目 系依的 及的入 法對輸 方比捷 入索快 輪檢到. 能爲達 智作而。 的來,述 新,下所 種伤形下 一部 清如 了母的明 共聲母説 {的韻細 音、詳 讀調之 字聲明 漢略發 用省本 使在關 參 青 第 圖 儲 與 碼 編行 進 音 .讀 字 漢 對 ,術 供技 提統 所傳 明於 發較 本相 據。 根珐 明方 說靳 以之 用存 丨訂---------0 p 一格 一規 4 I A S) N (C 準 標 家 國 國 中 用 適 度 尺 張 紙-本 527555 A7 B7 五、發明說明() (請先閱讀背面之注意事項再填寫本頁) 中,需使用4〜8個BYTE來儲存一個漢字讀音,本發明所使 用的編碼方式,僅使用了 2個BYTE( 16位元),便可表示一 個漢字的讀音。一般而言,注音符號可區分成作爲聲母使 用之“勹夂π匚力古了力《万厂Η <丁史彳尸0?古 修煩厶”,以及作爲韻母使用之“一 X u 丫石之廿巧 '幺只巧b jE* 讀· ίϊ 大厶儿,,。另外,藉著使闱“ 一聲(陰平)、二聲(〇、三聲 km (〆)、四聲(、)、輕聲(·),,,可表示漢字讀音之聲調。其 霉&中,由於一個漢字之讀音是由聲母、韻母或組合韻母(例 養夂如《乂石/ (國))以及聲調所組成。是以,如第一圖中所If you are troublesome, please win the f Μ department .: W ί :, Ping :; (I 33 Miscellaneous is mentioning 0 of the fifth, the description of the invention (After entering the phonetic symbol, you need to choose a word to get the required word In this way, the speed of input process using Zhuyin input method is also slower. In addition, because of Chinese Zhuyin input method, the operator needs to input the correct Mandarin phonetic symbol, so that the pronunciation of the input word is desired. When it ’s not easy to grasp, it often causes input errors. For example, roll tongues “force and EJ”, “i and down” and “corpse and 厶”, finals “Wang and Die”, “Qiao and You” and “~ and 厶” ", Or the Chinese tones are one, two, three, four, and soft, etc., which are easily confused by the operator, which leads to recognition errors and requires re-entry. Therefore, for phonetic not familiar with Mandarin Operators of the rules will find it inconvenient to use the traditional Chinese phonetic input method. Purpose and summary of the invention: (Please read the notes on the back before filling out this page) The vowel sound of the vowel in the article can be saved a little. Mention Entering the phonetic note of the present method \ lCul Species 1 for the presentation of the time sequence of the input process, as in the input of the middle note. Enter the system and the key method. Fang Jing first enters the first note. Note that the input number 0 in the input group 0 is read each time, and the input numbers for code numbers, note buys, and i injections are divided into zone readable, and the input numbers are described above. Transliteration of numbers: The mother's voice is printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. The coded coded sounds are read and entered into the two groups. Each coded root sound can be dealt with and 'next. Cheng Yin reads a check and marches in the same way to check this self-information. Words and words match the number of words and the mother and the mother's voice. Its coded coded tone reads the input, input, output, and output points. The plan can be followed in sequence. The paper size is applicable to the Chinese National Standard (CNS) A4 (210 x 297 g) 527555 A7 B7. Printed by the Consumers' Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. Yes No No arrears. 7 The word information of 0 is being mentioned by Yu Yue Qin; and then the word information is read from the address of the stored word information. Then, the simplified conditions for each set of input pronunciation codes can be determined to confirm the initials of the input pronunciation codes , Vowels, and simplifications of the tone. Based on the simplification conditions, each set of retrieval pronunciation codes of the word message is simplified. Then, compare whether the retrieval pronunciation code of the word message matches the input pronunciation code. When they match, then It is displayed on the screen for the operator to choose. The invention also provides an input method for inputting phrases and phrases. Among them, the phonetic symbols are first input, and the phonetic symbols are distinguished into array pronunciations, and then the retrieval process is performed according to the initials of each group of pronunciations in the array pronunciation, so as to select the word information covering the above initials combination from the database, where each A set of word messages covers the initials combination. Also, compare whether the pronunciation of the word message covers the input of the array. Then, all the word information covering the input pronunciation will be displayed on the screen according to the number of words for the operator to choose. Brief description of the drawings: The above description and the many advantages of this invention can be easily understood through the following detailed description combined with the attached drawings, where the first picture shows the pronunciation of Chinese characters according to the present invention. 1 6 Bit binary encoding and storage method; the second figure shows the database system for smart phonetic input method provided by the present invention; the third figure shows the main paper size of smart phonetic input method according to the present invention applicable to China National Standard (CNS) A4 Specification (210 X 297 Gt) ----------------- (Please read the unintentional matter on the back before filling this page) · 527555 A7 V. Description of the invention () The flow chart; the fourth diagram shows the use of Zhiyin Zhuyin input in accordance with the present invention to eliminate the use of each example; At 1 o’clock, the result displayed on the screen; The voice of the Zhuang voice shows that the fifth A ~ B picture shows the result displayed on the screen according to this post; the younger brother, and the picture show the process of meditation according to the present invention; short Sentence or slice into the phase of the French input phase. Shows an example of the application of the Yueyueguan Pass according to the present invention; ^ sentence or phrase J The eighth figure shows the related process of building a ~~, # #phrase according to the present invention;-The sound input method comes from • ih migration. The nine A ~ D pictures show the 3 red / Wang Yin input method ~~ results displayed on the screen when Huaiguang uses the self-built phrase according to the present invention; please read the precautions on the back before filling this page. The tenth figure shows the configuration provided according to the present invention. The input system • n ϋ nin. Check the detailed description: Timing issued by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs to print it for borrowing. 〇 According to the relevant entry method, the losing side was detected by the fast-track cable. It can work for Da Zhi. Come, talk about the new type of injury, and the next type of injury is as clear as the mother ’s. The consonant of the Ming dynasty said {the rhyme, the detailed pronunciation of the tune, and the statement of the Chinese version of Han Lufa used the provincial ambassador in Guan Shenqing's figure storage and code coding march. According to the facts, Jin Fang said that Jin Yizhi uses it to save and order --------- 0 p a Rule 1 IAS) N (C quasi-standard home country and middle school use moderate rule paper-this 527555 A7 B7 V. Description of the invention () (Please read the precautions on the back before filling this page), you need to use 4 ~ 8 Each BYTE is used to store a Chinese character pronunciation. The encoding method used in the present invention uses only 2 BYTE (16-bit) to represent the pronunciation of a Chinese character. Generally speaking, phonetic symbols can be distinguished as the initials used as "古 π 匚 Strengthen "WanchangΗ < Ding Shi 彳 Shi 0? Gu Xiu Tiao 厶" and "Yi X u 石 石 之 石 巧" used as a vowel, "biaoqiao b jE * read · ίϊ Ma'am, ... In addition, by making 闱 "one sound (yinping), two sounds (0, three km (〆), four sounds (,), soft sound (·),", the tone of the pronunciation of Chinese characters can be expressed. Its mildew & medium, Because the pronunciation of a Chinese character is composed of a consonant, a final or a combination of finals (such as Yang Yan, such as "乂 石 / (国)" and a tone. Therefore, as shown in the first picture

内 I 顯示,在本發明中,藉著使用第0〜5位元,來表示聲母; 是月 使用第6〜12位元來表示韻母;且使用第13〜15位元,來表示 聲調。 修所 正提 値得注意的是,對無聲母之漢字,例如“刃、(愛)” 。之 經濟部智慧財產局員工消費合作社印製 字而言,在聲母部份中亦給予對應於“無聲母”之編碼, 以顯示該漢字不具備聲母之情形。例如第一圖中顯示的 “ 00000 1 ”編碼,即用來代表該漢字聲母部份的狀態爲無 聲母。相對地,對於沒有韻母之漢字,例如“尸、(是)” 字,在韻母部份中亦以代表無韻母之“ 000000 1 ”來作爲 其編碼。此外,値得注意的是當具有組合韻母之情形發生 時,例如“ X e V/ (我)”字,則可將構成組合韻母之各個 韻母其編碼,加以組合而構成其相對應之組合韻母編碼。 例如上述“ X e〆(我)”字中之組合韻母“ X石”可以 使用代表“ X ”之編碼“ 0 1 0 0 0 0 0 ”與代表“冗”之編碼 “ 00000 1 1 ”之結合編碼“ 0 1 0 0 0 1 1 ”來表示。 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 527555The inner I shows that in the present invention, the initials are represented by using the 0th to 5th bits; the month is the finals using the 6th to 12th bits; and the 13th to 15th bits are used to express the tones. Xiusuo is paying attention to the Chinese characters of the silent mother, such as "blade, (love)". For characters printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, the initials are also given a code corresponding to "silent initials" to show that the Chinese characters do not have initials. For example, the "00000 1" code shown in the first figure is used to indicate that the initial state of the consonant of the Chinese character is the initial. In contrast, Chinese characters without finals, such as the word "corpse, (yes)", are also coded in the finals with "000000 1" that represents no finals. In addition, it is worth noting that when the situation with combined finals occurs, such as the word "X e V / (我)", the individual finals that compose the combined finals can be coded and combined to form their corresponding combined finals. coding. For example, the combination vowel "X shi" in the above "X e〆 (我)" word can use the combination of the code "0 1 0 0 0 0 0" that represents "X" and the code "00000 1 1" that represents "redundant" It is represented by the code "0 1 0 0 0 1 1". This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) 527555

五、發明說明() 煩請委員明示 > 年/月所提之#'正本>41、無變^^青發只内Igyxl。准」卞修正。_____^_ 經濟部智慧財產局員工消費合作社印製 如此一來,藉著第一圖 u γ耸母(罘0〜5位、)、韻母(第6〜12位)與聲調(第13〜15)之相捫姑) 3貝母(弟 '相關編碼,可表示出所選定的 漢字。例如,當輸入“電腦 一弓、 、”其聲母“为”可表爲000 1 1 0;而“ 母,由“ 一”與“弓”所共同 0011010;另外,其聲調爲“、” 可以得到整個“为—3 、 000 1 1 000 1 1 0 1 0 1 0 1,並可在㊉ ^ ............ 狂兒恥系統中以佔據2個BYTE 之編碼(0 X 1 8 D 5 )來加以表干 衣不。冋樣地,上述幺〆,,亦 可以 〇〇1〇〇〇〇〇〇1〇〇〇100之- 一、位編碼來表示,且在電腦系 統中,將其轉換爲使用兩個 1固ΒΥΤΕ<編碼(0x2044)來表 示。由此,藉著使用本發明所 十议月尸/Γ 供足編碼方式,將可大量 的節省儲存單元。 接著,請參照第二圖,該圖顯示了本發明中,操作 智能注音輸入法時,進行存取漢字所需之資料庫1〇〇。此 資料厍100可依據進行檢索之不同層次,而分爲一級索引 區102、二級索引區104、單字和片語信息區1〇6與單字或 片語信息組成區108。其中,一級索引區1〇2主要是根據輸 入字詞的罕數而進行區分。例如,可將輸入之字詞區分爲 單字、二字片語、三字片語…。在確認輸入字詞之字數後, 則可進入二級索引區丨〇4。此二級索引區i 04則是根據不同 的聲母組合而加以劃分。例如,當在一級索引區1 〇2中, 決定輸入字詞字數後,便可在二級索引區104内,檢索出 3 “ 1 由於其讀音爲“力 ,疋以可由第一圖中得知“为一巧 弓”爲組合韻 也成,其編碼可表示成 故可表示爲1 〇 1。由此, 字之二進位編碼爲 (請先閱讀背面之注意事項再填寫本頁) 裝V. Explanation of the invention () Members are kindly requested to indicate ># / 原本 mentioned in year / month > 41, no change ^^ Qingfa only Igyxl. Quasi "correction. _____ ^ _ This is printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. With the first picture u γ towering mother (罘 0 ~ 5th place), rhyme (6th ~ 12th place) and tone (13th ~ 15th place) ) 之 相 扪 姑) 3 Fritillaria (brother's related code, which can indicate the selected Chinese character. For example, when you enter "Computer Yi Gong,", its consonant "is" can be expressed as 000 1 1 0; and "mother, Common by "一" and "弓" 0011010; In addition, the tone of "," can get the entire "3-000 1 1 000 1 1 0 1 0 1 0 1 and can be found in ㊉ ^ .... ........ In the system of mad children, the code that occupies 2 BYTEs (0 X 1 8 D 5) is used to indicate whether it is dry or not. For example, the above, and also 〇〇〇〇 〇〇〇〇〇〇〇〇〇 100 of 100-one, the bit code to represent, and in a computer system, it is converted to use two 1-solid BΥTE < code (0x2044) to represent. Therefore, by using this The invention of the ten-day moon corpse / Γ foot-foot coding method can save a lot of storage units. Next, please refer to the second figure, which shows the operation of the intelligent phonetic input method in the present invention. At the time, the database 100 needed for accessing Chinese characters is 100. This data can be divided into a primary index area 102, a secondary index area 104, a single word and phrase information area 106 according to different levels of retrieval. And single-word or phrase information constitute area 108. Among them, the first-level index area 102 is mainly distinguished based on the number of input words. For example, the input words can be divided into single-word, two-word phrases, and three-word Phrase ... After confirming the number of words in the input word, you can enter the secondary index area. This secondary index area i 04 is divided according to different initial consonants. For example, when in the primary index area In 〇2, after determining the number of input words, 3 "1 can be retrieved in the secondary index area 104. Because its pronunciation is" force, "it can be known from the first picture that" is a coincidence bow " For the combination of rhyme, its code can be expressed as it can be expressed as 1 〇. Therefore, the binary code of the word is (Please read the precautions on the back before filling this page)

*11 I I 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 527555 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明說明( 修煩 正請* 11 I I This paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm) 527555 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs

II 内―厂til 疋州l/f 予E! 修所 正提 〇之 字數相同之聲母組合片語。亦即當所輸入之字詞經由一級 索引,確定輸入字詞有η個漢字後,可藉著二級索引檢索 出由η個聲母組成的片語。接著,根據二級索引之劃分, 可進入單字和片語信息區1 〇 6,以根據不同的聲母組合, 提供與聲母組合相符之片語信息。然後,再進入單字或片 語信息組成區1 0 8,以提取上述片語信息之讀音編碼。其 中,單字或片語信息組成區1 〇 8,即資料庫1 0 0中實際存放 單字或片語資料之位置。 値得注意的是,在電腦的實際操作中,往往經由加 密與壓縮過程,將讀音編碼(佔據了 2*N BYTE空間,其中 N爲該字詞的字數)轉換爲漢字加密碼,以進一步節省所 佔空間,並儲存於電腦系統中。是以在選定了某組單字或 片語,而進行提取其讀音編碼程序時,會先選擇其對應之 漢字加密碼,且經由解密與解壓縮過程,還原爲讀音碼。 在一實施例中,當輸入字詞爲“電腦”二字時,在 一級索引區1 02中,可將其劃分爲具有2個漢字之片語。接 著,在二級索引區1 04中,則可以聲母組合作爲區分依據, 而劃分出聲母組合爲“为”和“ 3 ”之片語區。然後,在 單字和片語信息區1 〇 6内,可根據上述聲母組合“为” “,”而選擇相對應的片語信息,例如“大腦”、“電 腦”、“當年”等等…。接著,在單字或片語信息組成區 10 8中,再儲存了將與上述片語信息(“大腦”、“電 腦”、“當年”…)對應之漢字加密碼,並可經由解密過 程而加以提取。 (請先閱讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 527555 A7 B7 五、發明說明() (請先閱讀背面之注意事項再填寫本頁) 接著,請參照第三圖,該圖所顯示爲根據本發明 所提供智能注音輸入法之主要流程。其+,首先進行步驟 2 0 0,輸入注音符號。亦即,經由操作者使用介面,將操 作者欲輸入漢字語詞之注音,輸入操作系統中。然後,進 ΐ具行步驟205,將所輸入之注音符號,區分爲複數組讀音。 示再進行步驟2 1 0,決定此複數組讀音中,每一組讀音的聲 黨¥母,如此將可得到對應於複數組讀音之複數個聲母。然 .¾ ^ 後,進行步驟215,自資料庫中選擇含括了此複數個聲母 9" ;'ν 之片語訊息。其中,此片語訊息是由分别對應於複數個聲 ψΐί 母的複數個漢字所組成。再進行步驟220,將所有片語訊 里^ 息之漢字讀音,與上述經區分的複數組讀音進行比較,以 ° ^ 決定所有片語訊息之漢字讀音,是否完全含括了複數組讀 音。接著,進行步驟225,顯示與上述複數組讀音相符合 之片語訊息,以供操作者選擇。要特别説明的是,由於整 個智能注音輸入法,在進行檢索比對的過程中,僅使用每 一假漢字讀音的聲母部份。是以,當操作者對某個漢字的 國語注音不甚確定時,僅需輸入此漢字之聲母,便可得到 具有此聲母之所有漢字。如此一來,即使操作者在輸入注 音時,省略了每個漢字讀音之韻母與聲調,但藉著本發明 所提供之智能注音輸入法,依舊可顯示出由這些聲母所構 成的可能片語,以提供操作者選擇。 經濟部智慧財產局員工消費合作社印製 至於智能注音輸入法之詳細流程,請參照第四圖。 該圖顯示了上述省略聲調與韻母之快捷輸入方式其應用 實例。例如,當操作者欲輸入電腦兩個字時,可經由如鍵 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐) 527555 A7 經濟部智慧財產局員 煩請委員明示年/2^.月所隄之 修正本有無變更實質内容是否准予修正。 五、發明說明( 300)。然後,签:入如、为1幺’,的注音符號(步驟 的注音符號,5 、輸入的注音加以區分’而決定出輸入 入注音爲“为二破分成多少組讀音(步驟3〇5)。例如 兩組讀音。:二時’可將其區分爲“力”和“… 字編碼方<,轉換/_分好的Λ音,根據本發明中新的漢 3 1 〇)。然後,、組1 6位兀的二進位讀音編碼(步線 再根據所產生二進i f 0 1 ::對第二_中之資料庫U1Q:二母信 出使用“为,,和“ 3 m 序。以讀 統中之儲存位η ^ #母的片語訊息,於資料庫系 入的注音特徵, 隧後根據铋作者輸 驟315中柃+ 疋母一組輸入讀音的簡化狀況,並對+ 铢甲k索到之片語訊息並為立 卫對步 (步驟320)。在太々丨士 /…貝9進仃相同的簡化程序 ^是本::广操作者所輸入之注音符號: 略了韻母與4 特徵乃:第一個讀音“为”: 調。是以;:二弟二個讀音…”則僅省略了聲 簡化私序,以去除片語訊息中 仃相同的 且去除第二個讀音的聲調。接著,…:::母與聲調, 碼與自資枓庠中經檢索與簡化程序二%乍:輸入的讀音 是否相符〜5)。在此例中,::: = =音碼 語却自鋒立 間化牙王序後的>5* “力;::、碼’與操作者所輸入之讀音碼(如此例中的 出的靖音碼相違:♦比對。當輸入的讀音碼與由資料庫中讀 心二則將具有此讀音碼之片語訊息,顯示 、勞幕上的候選區中’以供操作者選擇(步驟3。相反 (請先閱讀背面之注意事項再填寫本頁) 1*·^^ r * —1 n i mi i n m f l --- I if 111 1 MM·肅 · 消費 合 作 社 印 製 10Within II-factory til Luzhou l / f to E! The repair institute is mentioning the combination of initials with the same number of words. That is, when the input word is determined by the first-level index and there are n Chinese characters in the input word, the phrase consisting of n initials can be retrieved by the second-level index. Then, according to the classification of the secondary index, the single word and phrase information area 106 can be entered to provide phrase information corresponding to the initial combination according to different initial combinations. Then, enter the single word or phrase information composition area 108 to extract the pronunciation code of the above phrase information. Among them, the word or phrase information composes area 108, which is the location where the word or phrase data is actually stored in the database 100. It should be noted that in the actual operation of the computer, the encoding of the pronunciation (occupies 2 * N BYTE space, where N is the number of words) is often converted into Chinese characters and passwords through encryption and compression processes. Save space and store in computer system. Therefore, when a certain group of words or phrases is selected and the pronunciation encoding program is extracted, the corresponding Chinese character plus password is selected first, and the decryption and decompression processes are restored to the pronunciation code. In one embodiment, when the input word is "computer", it can be divided into phrases with two Chinese characters in the first-level index area 102. Next, in the secondary index area 104, the initial consonant combination can be used as a basis for distinguishing, and the phrase areas where the initial consonant combination is "for" and "3" are divided. Then, in the single word and phrase information area 106, the corresponding phrase information, such as "brain", "computer", "current year", etc., can be selected according to the above initials "for" ",". Then, in the single word or phrase information composition area 108, the Chinese characters corresponding to the above phrase information ("brain", "computer", "current year", ...) are stored and can be encrypted through the decryption process. extract. (Please read the precautions on the back before filling out this page) This paper size applies the Chinese National Standard (CNS) A4 (210 X 297 mm) 527555 A7 B7 V. Description of the invention () (Please read the precautions on the back before (Fill in this page) Next, please refer to the third figure, which shows the main flow of the smart phonetic input method provided by the present invention. Its +, first proceed to step 2 0 0, input the phonetic symbol. That is, the operator's use interface is used to input the phonetic notation of the operator's Chinese words into the operating system. Then, the process proceeds to step 205, and the input phonetic symbols are distinguished into plural array pronunciations. Step 2 10 is performed again to determine the sounds of each group of sounds in the complex array pronunciation. In this way, a plurality of initials corresponding to the sounds of the complex array can be obtained. Then, after ¾ ^, proceed to step 215, and select a phrase message including the plurality of initials 9 " 'ν from the database. The phrase message is composed of a plurality of Chinese characters corresponding to a plurality of sounds ψ 声 ί mother, respectively. Then, step 220 is performed to compare the Chinese character pronunciation of the ^ information in all the message with the distinguished complex array pronunciation, and determine whether the Chinese character pronunciation of all the phrase messages completely includes the complex array pronunciation by ° ^. Next, step 225 is performed to display a phrase message corresponding to the above-mentioned plural array pronunciations for the operator to choose. It should be particularly noted that, because of the entire intelligent Zhuyin input method, only the initial part of each fake Chinese character pronunciation is used in the search and comparison process. Therefore, when the operator's phonetic notation of a certain Chinese character is not certain, just input the initial of the Chinese character to get all the Chinese characters with this initial. In this way, even if the operator omits the vowels and tones of each Chinese character when inputting the phonetic notation, the intelligent phonetic input method provided by the present invention can still display possible phrases composed of these initials. To provide operator choice. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs As for the detailed process of the smart phonetic input method, please refer to the fourth figure. This figure shows an example of the application of the above-mentioned shortcut input method omitting tones and finals. For example, when the operator wants to input two characters of the computer, the paper size of the keypad can be applied to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 527555 A7 The member of the Intellectual Property Bureau of the Ministry of Economics, please clarify the year / 2 ^ . Whether there is any change to the substance of the amendment of Yuesuo Dike? 5. Description of the invention (300). Then, sign in: the phonetic symbol of 1 、 ', (the phonetic symbol of the step, 5 and the input phonetic to distinguish the', and decide the input phonetic is "how many pronunciations are divided into two (step 305) For example, two sets of pronunciations: "two o'clock" can be distinguished into "force" and "... word encoding method < conversion / _ good Λ sound, according to the new Chinese 3 1 0). Then, group 1 6-bit binary phonetic encoding (step line is based on the generated binary if 0 1 :: for the database U1Q in the second _: the second mother letter out uses "为 ,," and "3 m Preface. Based on the phonetic information of the memory bit η ^ # in the reading system, the phonetic features incorporated in the database. After the tunneling, according to the simplified state of the input pronunciation of the group 柃 + 疋 in the bismuth author input 315, and + The phrase message obtained by Kazakhstan K and stepped back for Li Wei (step 320). Enter the same simplified procedure in Taiji, Shi / ... Bai 9 ^ This is the phonetic symbol entered by the operator: The omitted vowels and 4 features are: the first pronunciation "is": tune. Is ;: two brothers two pronunciations ... "only omitted the voice to simplify the private order, in order to remove the same in the phrase message and remove the first The tone of the two pronunciations. Then,… ::: mother and tone, code and self-funded search and simplification procedure 2%: whether the input pronunciation matches ~ 5). In this example, ::: = = Phonetic code but since the order of the frontal teeth of the King of Tooth> 5 * "force; ::, code 'and the phonetic code input by the operator (as in this example, Jingyin code conflicts: ♦ Compare. When the entered phonetic code is read from the database, the phrase message with this phonetic code will be displayed and displayed in the candidate area on the screen for the operator to choose (step 3. Instead (please read the precautions on the back before filling this page) 1 * · ^^ r * —1 ni mi inmfl --- I if 111 1 MM · SU · Printed by Consumer Cooperative 10

527555 經濟部智慧財產局員工消費合作社印製 煩請委員明示^一年h月所提之 修正本有無變更實質内容是否准予修正。 A7 B7 五、發明說明( 的,當所輸入的讀音碼,與由資料庫中片語訊息所簡化的 讀音碼不相同時,則跳過此片語訊息,且不顯示於螢幕 上。接著,判斷所有片語訊息是否已處理完畢(步骤3 3 5 ), 若尚未處理完畢,則再跳回至步驟3 2 0,重新比對操作者 輸入讀音碼與步驟3 1 5中所檢索到之下一個片語訊息。 是以當操作者輸入“为7幺”時,螢幕上將可出現 如第五A圖中所顯示之結果。其中,當智能注音輸入系統 根據上述流程進行檢索選字時,可將包括1 .電腦,2 ·打 鬧,3 .大腦…等含括“为3幺”注音之可能片語訊息,全 部顯示於螢幕上,以供操作者進行選擇。相對的,當操作 者僅輸入“为3 ”時,智能注音輸入系統亦會顯示如第五 B圖中,包括1 .當年、2 .電腦、3 .大腦、4 .叮嚀、5 ·打鬧… 等對應之片語訊息,以供操作者選擇。 値得注意的是本發明所提供之智能注音輸入法,除 了可根據操作者輸入之注音符號,在決定其讀音數目後, 提供與每一組讀音其聲母相符合之字詞外,亦可特别針對 短句或片語之輸入。請參照第六圖,該圖顯示應用本發明 方法於短句或片語輸入之詳細流程。首先,經由鍵盤輸入 注音符號(步驟400)。然後,如同上述,藉著將輸入的注 音加以區分,而決定區分爲多少組讀音(步驟4 0 5)。接著, 將區分好的注音,轉換爲1 6位元的二進位編碼(步驟 4 1 0)。然後,再根據所產生二進位編碼中之聲母信息,對 第二圖中之資料庫系統1 0 0,進行檢索程序,以讀出包含 上述聲母信息,且具有不同長度之所有片語、短句,在資 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) (請先閱讀背面之注意事項再填寫本頁)527555 Printed by the Employees' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs Members are kindly requested to indicate ^ whether the amendments mentioned in the previous year are subject to change. A7 B7 5. Description of the invention (, when the input pronunciation code is different from the simplified pronunciation code of the phrase information in the database, the phrase message is skipped and not displayed on the screen. Then, Determine whether all the phrase messages have been processed (step 3 3 5), if not, then skip back to step 3 2 0 and re-compare the operator to enter the pronunciation code and retrieved from step 3 1 5 A phrase message. When the operator inputs "is 7 幺", the result shown in Figure 5A will appear on the screen. Among them, when the intelligent Zhuyin input system searches for words according to the above process, Possible phrase messages including 1. computer, 2. harassment, 3. brain, etc., including "for 3“ "phonetic, all displayed on the screen for the operator to choose. In contrast, when the operator When only “is 3” is input, the smart phonetic input system will also display the corresponding phrase information as shown in Figure 5B, including 1. current year, 2. computer, 3. brain, 4. ding dong, 5. • noisy ... For the operator's choice. It should be noted that the present invention The intelligent phonetic input method provided can not only provide the words corresponding to the initials of each group of pronunciation after determining the number of pronunciations according to the phonetic symbols input by the operator, but also can specifically target the input of short sentences or phrases. Please refer to the sixth figure, which shows the detailed flow of applying the method of the present invention to short sentence or phrase input. First, input the phonetic symbol through the keyboard (step 400). Then, as above, distinguish the input phonetic notation , And decide how many groups of pronunciations are distinguished (step 4 5). Next, the distinguished phonetic notation is converted into a 16-bit binary code (step 4 1 0). Then, according to the generated binary code, The initial information of the initials is searched on the database system 100 in the second figure to read out all the phrases and phrases containing the above initials and having different lengths. The Chinese national standard is applied at the capital paper scale ( CNS) A4 size (210 X 297mm t) (Please read the precautions on the back before filling this page)

527555 修煩 主請 本委 無明 一了,0 秦所 主提 〇之 經濟部智慧財產局員工消費合作社印製 A7 五、發明說明() 2厍系統中之儲存位址與數量(步驟415)。隨後,將檢索 =片語、短句,根據其不同長度,由長至短依次進行劃 刀人驟420)。再根據掭作者輸入注音的特徵,將步驟 中檢索到之資料,進行相同的簡化程序, 輸入注音進行比對(步—接著,比對所輸 碼過檢索與簡化程序後之片語訊息讀音碼是否相同 :κ 43 0)。其中,値得注意的是,當輸入之讀音碼具有N 個項音時,祇需比對檢索片語訊息的前_讀音碼是否相 同即可,是以檢索片語訊息所具有的讀音碼數目(即其字 數)亦可大於N。當輸入之讀音碼與片語訊息簡化後的 1買音碼相同時,可將檢索出來的片語訊息,顯示於答幕上 之候選區中,供操作者選用(步驟43 5)。接著,決定目前 所處理具有相同字數之片語、短㈣息是否'已全部處埋* 畢(步驟440),若尚未處理完畢,則再跳回至步驟Ο。= 已將具有同樣字數之片語、短句信息處理完畢,則針對字 數較少之片語、短句信息繼續進行處理(步驟445), 可回跳至步驟420,以處理詞語信息。當所處理之片語、 短句其1買音數目小於操作者輸入讀音碼數目 理程序。 處 請參照第七圖’該圖顯示了使用本發明方法,輪入 片語、短句模式之相關實例。其中,當操作者輸入 經上述步驟405,可將其區分爲“ “ 了” 、 尸一個且其簡化特徵同樣爲省略了逝 調。是以,在對資料庫進行檢索後,可將所讀取具有^ 12 本紙張尺度適用中國國家標準(CNS)A4規格(210 (請先閱讀背面之注意事項再填寫本頁)527555 Repair trouble, please ask this committee. I have no idea. 0 Qin Institute. 〇 Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. A. 5. Description of the invention (2) Storage address and quantity in the system (step 415). Then, search = phrase, short sentence, and according to its different length, from the longest to the shortest, perform the scribe step 420). Then according to the characteristics of the input phonetic transcription of the author, the data retrieved in the step is subjected to the same simplified procedure, and the phonetic transcription is input for comparison (step-then, the coded phonetic code after searching and simplifying the input code is compared) Is it the same: κ 43 0). Among them, it should be noted that when the input pronunciation code has N items, it is only necessary to compare whether the first _ pronunciation code of the retrieval phrase message is the same, and the number of pronunciation codes of the retrieval phrase message can be compared. (That is, its number of words) can also be greater than N. When the input phonetic code is the same as the simplified 1-buy code of the phrase message, the retrieved phrase message can be displayed in the candidate area on the answer screen for the operator to choose (step 43 5). Next, it is determined whether the phrases and short messages currently processed with the same number of words have been “buried” (step 440). If the processing has not been completed, then skip back to step 0. = The phrase and phrase information with the same number of words has been processed, then the phrase and phrase information with fewer words will continue to be processed (step 445), and you can skip to step 420 to process the word information. When the phrase or phrase processed is less than the number of phonetic codes entered by the operator, the number of sounds is 1. Please refer to the seventh figure. This figure shows a related example of the turn-by-pass phrase and short sentence pattern using the method of the present invention. Among them, when the operator inputs through the above step 405, it can be divided into "" 了 ", corpse one, and its simplified feature also omits the deviance. Therefore, after searching the database, it can be read With ^ 12 This paper size applies to China National Standard (CNS) A4 specifications (210 (Please read the precautions on the back before filling this page)

527555 、 Α7 Β7 五、發明說明( 修煩 正請 f員 十右明 f示 ψ 冬(1 余所 正提 〇之 經濟部智慧財產局員工消費合作社印製 (請先閒讀背面之注意事項再填寫本頁) 字數之片語、短句,在經過相同的簡化程序後(此例中, 將第一至第三組讀音去除韻母與聲調),與所輸入讀音碼 進行比對。例如,輸入“ < 7尸”時,勞幕上將會出現“請 您使用智能注音輸入法”之短句。並且,藉著移動滑鼠按 壓往右之箭頭,可選擇其它符合之短句與片語。 此外,本發明所提供之智能注音輸入法,亦具有自 建片語之功能,以便增加資料庫中儲存的資料。請參照第 八圖,該圖顯示自建片語之相關流程。其中,操作者可經 由鍵盤輸入注音符號(5 00)。同樣的,對所輸入的注音符 號進行區分讀音與轉碼之程序,而產生了複數組讀音碼 (5 0 5)。然後,可按照輸入讀音的順序,選擇候選字詞, 並根據操作者所選定的字詞,紀錄此字詞完全的讀音 (5 10),直到自建程序完畢爲止。如同第九A圖至第九C圖 所示,其中當操作者輸入“下丁us Η —丫力一巧”時, 智能注音系統會針對輸入讀音碼,依照讀音順序,以片語 或單字顯示,並供操作者選擇。如第九Α圖中,顯示了代 表“下丁 U b ”讀音碼之“資訊”。然後,在第九B圖 中,則依序顯示“ Η — 丫” (家),且在第九c圖中,顯示 了“为一巧”(電)字。如此,可依序完成選字程序,並同 時記錄了“資訊家電”中每一個漢字的完全讀音。接著, 在完成自建片語後,可根據自建片語的前兩個聲母,與片 語的字數,而定位出其於資料庫中應存放之位置(5 1 5)。 再檢查該位置處,以確認是否已有與新建片語相同的詞句 存在(5 2 0 ),當資料庫中並不存在相同的詞句時,則將新 13 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) 527555527555, Α7 Β7 V. Description of the invention (If you are in trouble, please ask the staff member to show your name and show ψ winter (printed by more than 1 of the Ministry of Economic Affairs Intellectual Property Bureau employee consumer cooperative) (please read the precautions on the back first) Fill in this page) After the same simplified procedure (in this example, removing the vowels and tones from the first to third sets of pronunciations), compare them with the input pronunciation code. For example, When you enter "< 7 Corpses", the phrase "Please use the intelligent Zhuyin input method" will appear on the labor screen. And, by moving the mouse and pressing the right arrow, you can select other matching phrases and films. In addition, the intelligent Zhuyin input method provided by the present invention also has the function of self-created phrases in order to increase the data stored in the database. Please refer to the eighth figure, which shows the related process of self-created phrases. , The operator can input Zhuyin symbols (5 00) through the keyboard. Similarly, the input phonetic symbols are distinguished from pronunciation and transcoding procedures to generate a complex array of phonetic codes (5 0 5). Then, according to the input Pronunciation order Select a candidate word and record the complete pronunciation of the word (5 10) according to the word selected by the operator until the self-built procedure is completed. As shown in Figures 9A to 9C, when the operation When the user enters “下 丁 us Η — 力 力 一 巧”, the intelligent phonetic transcription system will display the input pronunciation code in the phrase or word according to the pronunciation order, and it is for the operator to choose. As shown in Figure 9A, Represents the "information" of "Xia Ding U b" pronunciation code. Then, in Figure 9B, "图 中-丫" (home) is displayed in sequence, and in Figure 9c, "is a coincidence" "(电)". In this way, the character selection process can be completed in sequence, and the complete pronunciation of each Chinese character in "Information Appliances" is recorded at the same time. Then, after completing the self-built phrase, you can The two initials, and the number of words in the phrase, locate the location where it should be stored in the database (5 1 5). Then check the location to confirm whether the same phrase as the newly created phrase exists (5 2 0), when the same phrase does not exist in the database, the new 1 3 This paper size is applicable to China National Standard (CNS) A4 (210 X 297 g) 527555

AT B7 五、發明說明( i'l· ,!ΐ:! 5 二:-丨ill 辦 U ,f/.L π· 經濟部智慧財產局員工消費合作社印製 建好的片語,連同其完全的讀音碼皆插入資 位置⑽)。同時,針對資料庫⑽中之:級索引區i㈣ 行修改,以更改其起始位址與詞語數量(53〇)。如此一來, 當操作者於鍵盤上輪入“川勿”日争,智能注音系统將 可纟二由上述檢索流程,而出現如第九D圖中“資訊家電” 之螢幕畫面。 / " 請參照第十圖,#圖顯示了本發明所提供之智能注 音輸:系統6〇〇。其中,此系統具有-輸入裝置610,用以 H # # # it彳t & f輸入程序。並且’—讀音 620,回應於輸入裝置61〇,用以將操作者所輸入之注音, 區分爲數組讀音,且決定此數組讀音中,每一組讀音所包 含的聲母。另夕卜’―個二進位轉碼器63q,連接至該讀音 拆分裝置620,用以將_產每 > 姑 、± 、 肝所產生之數組讀音,轉換爲複數組 16位π之二進位編碼。値得注意的是,每一個漢字皆可以 第二圖中所顯示的方法’使用2個byte的二進位編碼來加 '表丁 ‘、、:後 私索系統6 4 0,連接於一資料庫1 〇 〇且回 應於上述二進位轉碼器63 0’以便根據所輸入之讀音碼, 進行索引程序。如此’ τ在資料庫刚中,找到與輸入讀 音碼對應的起始位址與資料數量。並且,—簡化比對系統 650,用以決定所輸入讀音碼之簡化特徵,並將檢索系統 6 4 0由資料庫i 〇 〇中所取得之資料,以相同的簡化程序加以 簡化,再與所輸入的讀音碼進行比較。其中,冑簡化後的 檢索讀音碼,與由操作者所輸入之讀音瑪相符時,則可將 此筆檢索讀音碼所代表之詞句,&由—顯示裝置“Ο加以 (請先閱讀背面之注意事項再填寫本頁) 裝--------訂----AT B7 V. Description of the invention (i'l ·,! Ϊ́ :! 5 2:-丨 ill Do U, f / .L π · The printed phrase printed by the employee consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs, together with its complete The pronunciation codes are inserted into the position ⑽). At the same time, the database: 索引: level index area i㈣ is modified to change its starting address and the number of words (53). In this way, when the operator enters the "Chuan Wu" day contest on the keyboard, the smart phonetic system will be able to go through the above retrieval process, and the screen of "Information Appliances" shown in Figure 9D will appear. / " Please refer to the tenth figure. The # figure shows the intelligent phonetic input provided by the present invention: the system 600. Among them, the system has an input device 610 for inputting a program with H # # # it 彳 t & f. And'—phonetic 620, in response to the input device 61, is used to distinguish the phonetic input by the operator into an array of phonetic sounds, and to determine the initials contained in each group of phonetic sounds in this array of phonetic sounds. In addition, a binary transcoder 63q is connected to the pronunciation splitting device 620, and is used to convert the array readings produced by _____, ±, and liver into a complex array of 16-bit π / 2 Carry encoding. It should be noted that each Chinese character can be added to the table using the 2-byte binary encoding method shown in the second figure: "Post Ding", :: The private search system 6 4 0, connected to a database 1 00 and in response to the above-mentioned binary transcoder 63 0 'in order to perform an indexing process based on the input pronunciation code. In this way, τ found the starting address and data quantity corresponding to the input phonetic code in the database. Moreover, the simplified comparison system 650 is used to determine the simplified features of the input phonetic code, and the information obtained by the retrieval system 6 40 from the database i 00 is simplified by the same simplified procedure, and then simplified with The entered phonetic codes are compared. Among them, when the simplified retrieval pronunciation code is consistent with the pronunciation entered by the operator, the words and sentences represented by the retrieval pronunciation code can be obtained from the display device "〇 (Please read the first (Please fill in this page for attention) Install -------- Order ----

P 14 527555 Α7 Β7 顯示,以提供操作者選用 煩請委蓮嗔命^^年/^月^!11听提之 修正木奄#變更實質内容是否准予修正。 ___ 經濟部智慧財產局員工消費合作社印製 五、發明說明( 合於簡化比對系統650與資料庫子3擴无模組670輕 貝竹/皁1 〇 0,用以進行丘 序,以增加資料庫中之片語資料。 °° 本發明所提供之智能注音輸 點。其中,藉著使用漢定讀音之聲母部份,來作爲二: 對的依據,可以省略聲調、韻母而達到快捷輸入的目的。 例如,不管鍵入“史X z “ ” 八u m 、“史〈〈χ石” “史 人《”或是“虫《”,皆會在螢幕上出現“中國 漢字的選項。同樣地,只要輸入“为了”,便能夠輸入 H 一來’將可大幅縮減輸入韻母與聲調所耗費的 時間,而有效的提高輸入效率。同時,也可減少由於不孰 悉注音規則所導致的時間浪費。另外,智能注音輸入法… 並具有片5吾、短句自動匹配之功能。其中,在輸 短句的模式中,僅需鍵入少數幾個注音,便可輸入:: 長的片浯或短句。例如,只要鍵入“ 乂 “黄河之水天上來,,之短句。因爲厂“厂厂气輸入 “ ± ”分别是這段文字開頭三個纟的聲母。如此—厂來,^ 然可大幅縮短成批輸入的時間。更者,由於智能注立”’、、 並具有自建片語的功能,是以可提供操作者自已::: 語,而達到擴充詞庫,方便日後輸入之功能。 k 本發明雖以一較佳實例闡明如上,然其並 定本發明精神與發明實體,僅止於此一實施例爾。對:: 此領域技藝者,在;F㈣本發明之料與範圍内:二 改,均應包含在下述之申請專利範圍内。 ‘ 15 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公餐y -------I--- I --- (請先閱讀背面之注意事項再填寫本頁) .P 14 527555 Α7 Β7 is displayed to provide the operator with a choice. I would like to ask you to order your life ^^ year / ^ month ^! 11 Listen to it. Modify the wood ## Whether the substance of the change is allowed to be revised. ___ Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. 5. Description of the invention (combined with the simplified comparison system 650 and the database 3 expansion module 670 light shell bamboo / soap 1 00, used to perform the sequence, to increase Phrase data in the database. °° The intelligent phonetic input point provided by the present invention. Among them, by using the initial part of the Handing pronunciation, as the second: the basis of the right, you can omit the tone and vowel to achieve quick input For example, no matter if you type "historical X z" "eight um," history << χ 石 "" historical man "or" worm ", the option of" Chinese characters "will appear on the screen. Similarly, As long as you enter "for", you can enter H once to greatly reduce the time it takes to enter finals and tones, and effectively improve the input efficiency. At the same time, it can also reduce the time waste caused by not knowing the phonetic rules. In addition, the intelligent phonetic input method ... also has the function of automatic matching of short sentences and short sentences. Among them, in the mode of inputting short sentences, you only need to type a few phonetic transcriptions, and you can enter: For example, as long as you type "黄", the water of the Yellow River is coming up, and the short sentence. Because the factory "factory gas input" ± "are the initial consonants of the three 纟 at the beginning of this text. So—the factory, ^ can be greatly shortened The time for batch input. In addition, due to the intelligent injection "'," and the function of self-built phrase, it can provide the operator's own :: language to expand the thesaurus and facilitate future input. k Although the present invention is explained as above with a preferred example, it also defines the spirit and the inventive entity of the present invention, and it is limited to this embodiment. To :: Those skilled in the art, within the scope and scope of the present invention: The second amendment should be included in the scope of the patent application below. '15 This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 public meals y ------- I --- I --- (Please read the notes on the back before filling this page).

Claims (1)

527555 A8 B8 C8 D8 申請專利範圍 列 下 括 包 少 至 法 方 該 法 方 入 輸 音 注 能 智 種 驟 步 煩請委員明示片年d ”所,提之 修正本有無變更實質内容是否准予修正〇 ;一母 號少聲 符至之 音 注 個 1 少 至 入 輸 ,,定 音決 讀而 擇 選 中 庫 料 資 爲音從 號讀以 符該序 音組程 注一索 該每檢 分據行 區根進 合 .,組 合母 組聲 母該 聲括 - 含 自心 訊 詞 字之 且,。 音擇 讀選 組供 各以 該息 括訊 含詞 否字 是該 音之 讀音 之讀 息組 訊各 詞該 字蓋 該涵 較示 比顯 至 該 分 區 在 中 其 法 方 之 項 11 第 圍 範 利 專 請 中 如 2 少 至 該 將 括 包 〇 更碼 後編 音音 士買 士買 --0.--tl 組位 一進 少二 至組 爲一 號少 符至 音成 注換 個轉 一 音 少讀 經濟部智慧財產局員工消費合作社印製 3 .如申請專利範圍第1項之方法,其中上述之資料 庫包括: 一級索引區,根據該至少一組讀音而劃分;二級索引區’回應於該一級索引區,且根據該聲母組 合而劃分; 單字和片語訊息區,回應於該二級索引區,用以提供 與該聲母組合相符之字詞訊息;及單字或片語訊息組成區,用以存放該字詞訊息之讀音 編碼。 I J----------裝·-------訂·-------- (請先-¾讀背面之注意事項再填寫本頁) 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公釐)527555 A8 B8 C8 D8 The scope of the patent application includes as little as the French party. The French party can enter the phonetic notation, and the members are kindly asked to indicate the year d ", whether there is any amendment to the amendment, and whether the substance is allowed to be amended. A mother's note with a note and a note of 1 should be entered as low as one, and the final note must be selected and selected. The library materials should be read as a note. The note should be the same as the sequence of notes. Ougen Jinhe., The combination of the initial consonants of the consonant-including words from the heart of the heart, and. The sound selection reads the group for each reading of the message containing the word whether the word is the pronunciation of the sound Each word covered by this word is displayed to the extent that it is displayed in the section of the French method in the 11th paragraph Fan Li special request, such as 2 as low as it should be included 〇 Recoded sounds buy buy--0 .-- tl The group is entered one less and the second is grouped with the number one lesser and the note is changed. One is changed and the other is read. The number is printed by the employee consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs. The above database includes: The index area is divided according to the at least one set of pronunciations; the secondary index area is responded to the primary index area and divided according to the initial combination; the word and phrase information area is responded to the secondary index area for providing A word message that matches the initials; and a single word or phrase message composition area, used to store the pronunciation code of the word message. I J ---------- installation ------- -Order · -------- (Please read the notes on the back before filling in this page) This paper size is applicable to China National Standard (CNS) A4 (210 X 297 mm) ABCD 經濟部智慧財產局貝工消費合作社印製 '令請專利範圍 4·如申請專利範園苐1項之方珐,其中比較該字詞 訊息之讀音與該至少一組讀音之步驟,更包括下列步驟: 決定該至少一組讀音之簡化程序,以確認該每一組 讀音其聲母、韻母與聲調之簡化狀況; 根據該簡化程序,將該字詞訊息之每一組讀音簡 化,而取得該字詞訊息之簡化讀音;且 比較該字詞訊息之簡化讀音是否涵蓋該至少一組 讀音。 5· —種使用智能注音輸入法來進行自建片語庫之 程序,該程序至少包含下列步驟: 輸入欲增加片語之注音符號; 區分該注音符號爲至少一個讀音· 根據該至少一個讀音之順序,依次提供候選字詞以 供確認,而產生一組片語字詞; 決定該組片語字詞於該資料庫中存放之位置丨且 儲存該組片語字詞於該資料庫中。 ’ 6· Μ請專利1^第5項之程序,其中在 :?字詞之前’更包括進行-檢查步味,以檢查該存放位 置疋否已有與該組片語字詞相同之詞句存在。 7·如申請專利範園第5項之程庠 片这含叫接,#勺扭4 &gt; •序,其中在儲存該組 Π谙竽W後,更包括進行一更改 ^ 更改程序,以更改該資料庫中 17 F /ΌΧΤ〇\ , »..............裝.........訂.........^ &lt;請先閲讀背面之注意事項再填窝本頁} 527555 A8 B8 C8 D8 申請專利範圍 索引區之起始位址與詞語數量 修煩ftt嘗驟: 無明 _示 質I 8. 種智能注音輸入方法,該方法至少包括下列步 輸入 疋六]|/f 予d 修所 正提勹44-。之包栝 輸入至少一組注音符號; 區分該注音符號爲至少一組讀音; 轉換該每一組讀音爲輸入讀音編碼,其中每一組該 讀音編碼均爲二進位讀音編碼; 根據每一組該輸入讀音編碼之聲母,形成聲母組合; 根據該聲母組合,進行檢索程序,其中檢索程序更 (請先閱讀背面之注意事項再填寫本頁) -裝 經濟部智慧財產局員工消費合作社印製 進行第一檢索程序 數目相同之字詞訊息; 進行第二檢索程序 相符合之字詞訊息;且 自儲存該字詞訊息之位址,讀取該字詞訊 息,其中每一組該字詞訊息均涵蓋該聲母組合; 決定每一組該輸入讀音編碼之簡化程序,以確認該 輸入讀音編碼其聲母、韻母與聲調之簡化狀況; 根據該簡化程序,對該字詞訊息之每一組檢索讀音 編碼進行簡化; 比較該字詞訊息之檢索讀音編碼與各組該輸入讀 音編碼是否相符合;且 顯示與各組該輸入讀音編碼相符合之該字詞訊 以劃分出與該聲母組合 以劃分出與該聲母組合 * I I I # 18 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) 527555 A8 B8 C8 D8 申請專利範圍 自A 9 修正本有無變更實質内容是否准予修正。 煩請委員明示% ^ 庫 吾 、一&quot; 片 含 包 更 中 其 法 方 g 之d號個序 項々符一順 8少音少之 第至 1 f . 注至音 圍序 範程 利該 專, 請序 申程 如之 之爲讀 語號個 片符 一 加音少 增注至 欲該該 入分據 輸區根 確 者 作 操 生 產 以 驟 步 列 下 含 供 詞 字 選 候 供 提 ;次 音依 讀 ’ 詞 字 語 片 且 置 位 之 放 存。 中中 庫庫 料料 資資 該該 於於 詞詞 字字 語語 片片 1cd &quot;0 該該 定存 決儲 Θ所提之 第 圍 範 利 專 請 申 如 組位 該放 存存 儲該 在查 巾檢I 其以 ,,存 法驟句 方步詞 之查之 項檢同 :一相 一丁 )1¾ 進字 括語 包片 更組 , 該 前與 之有 詞已 字否 語是 片置 # 二 青包i 更位t,始 i後起 1詞之 1 字區 語引 片索 第 圍 範 利 組_ 該庫 存料 儲資 在該 中改 其更 ; 以 法, 方序 之程 項改。 :更量一數 行語 進詞 I -----------裝·-------訂 *-------- (請先閱讀背面之注意事項再填寫本頁) 智 , 種置 一裝 .入 12輸 經濟部智慧財產局員工消費合作社印製 裝 分 拆 音 讀 括序 包程 少入 至輸 統音 系注 該行 ,進 統者 系作 入操 輸供 音提 注以, 能用置 音 讀 組 ; 一母 少聲 至組 爲一 音少 注至 入之 輸含 分包 區所 以音 ΟΠ 士買 月-ηα ,組 置一 裝少 入至 輸該 該出 於分 應區 回且 9 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t ) 527555 A8 B8 C8 D8 申請專利範圍 裝少 分至 拆之 音母 讀聲 該組 於 一 應少 回至 且該 庫於 料應 資對 一取 於讀 接中 連庫 ,料 統資 系該 索由 檢便 以 置 煩請委員明示St年/1月 修正本有無變更實質内容是否准予修正 詞 字 個 示 顯 之 索 檢 經 示 顯 以 用 統 系 索 檢 該 於 應 回 置 且裝 擇 選 與 認 辨 行 進 者 作 操 該 供 提 以 詞 字 個 1 少 至 該 一 索 。 括檢碼 包該編 更於音 中 合 讀 其耦的 ,且位 、无,進 ;N 一一 之裝爲 ¾分換 2 拆轉 11 第音音 圍讀讀 範該組 利於 一 專應少 請回至 申 該 如器將 .碼以 3 轉用 1位, 進統 二系 統 系 對 比 化 簡 lal所提之 4 如 一 且 括, 包統 更系 中索 其檢 ,該 統與 系器 之石. 項轉 第進 圍二 範該 利於 專應 請回 中 且與 , 再 件, 條化 化簡 簡以 之加 碼件 編條 音化 讀簡 該該 定以 決詞 以字 用之。 ,取對 置讀比 裝所行 示統進 顯系碼 該索編 於檢音 合該讀 耦將該 模 充 擴 詞 字 一料 括資 包該 更於 中合 其耦 , 且 統, 系統 之系 項對 2 比 11 第化 圍簡 範該 利於 專應 請 如 回 資 語 片 之 中 庫 料 資 加 增 以 序 程 語 片 g-c. 自 行 進 以 用 «/ Ο 庫料 (請先閱讀背面之注意事項再填寫本頁) ▼裝 • m HI —I In n n vm 一 t ft m m In n # 經濟部智慧財產局員工消費合作社印製 20 本紙張尺度適用中國國家標準(CNS)A4規格(210 X 297公t )ABCD printed by the Intellectual Property Bureau of the Ministry of Economic Affairs of the Bayer Consumer Cooperatives. Scope of patents 4. If you apply for a patent, the first paragraph of the square enamel, in which the steps of comparing the pronunciation of the word message with the at least one group of pronunciation, including The following steps: Determine the simplified procedure of the at least one set of pronunciations to confirm the simplified state of the initials, finals and tones of each set of pronunciations; according to the simplified procedure, simplify each set of pronunciations of the word information to obtain the The simplified pronunciation of the word message; and comparing whether the simplified pronunciation of the word message covers the at least one set of pronunciations. 5. · A program for self-built phrase library using intelligent Zhuyin input method, the program includes at least the following steps: input the phonetic symbol of the phrase to be added; distinguish the phonetic symbol as at least one pronunciation · According to the at least one pronunciation In order, a set of phrase words is provided in order for confirmation, and a set of phrase words is determined; the position of the set of phrase words in the database is determined, and the set of phrase words is stored in the database. '6 · M, please apply for the procedure of item 1 ^ 5 of the patent, in which: before "? Words" also includes the process of-check the step taste to check whether the storage location already has the same phrase as the phrase phrase . 7 · If the process of applying for the patent No. 5 in the park, this is called 接 接, # spoon twist 4 &gt; • sequence, after storing the group Π 谙 竽 W, it also includes making a change ^ Change the program to change 17 F / ΌΧΤ〇 \ in this database, ».............. install ......... order ......... ^ &lt; please Read the precautions on the back before filling in this page} 527555 A8 B8 C8 D8 The starting address and the number of words in the index area of the patent application scope Fft tasting: Wuming_Quality I 8. A smart phonetic input method, which The method includes at least the following steps. Enter 疋 六] | / f to d. Include at least one group of phonetic symbols; distinguish the phonetic symbols into at least one group of pronunciations; convert each group of phonetic transcriptions into input pronunciation codes, where each group of the pronunciation codes are binary phonetic codes; according to each group of the Enter the initials of the pronunciation code to form a consonant combination. According to the consonant combination, a search procedure is performed, and the search procedure is even more (please read the precautions on the back before filling this page). A word message with the same number of search procedures; a word message corresponding to the second search procedure; and read the word message from the address where the word message is stored, where each group of the word message covers The initial combination; determine the simplified procedure of each input pronunciation code to confirm the simplified status of the initial pronunciation, finals and tones of the input pronunciation code; according to the simplified procedure, perform the retrieval pronunciation code of each group of the word information Simplify; compare the search pronunciation code of the word message with the input pronunciation code of each group; and display The word and message corresponding to the input pronunciation code of each group is divided into a combination with the initials to divide into a combination with the initials * III # 18 This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) ) 527555 A8 B8 C8 D8 The scope of patent application is from the amendment of A 9 whether the substance of the amendment is allowed to be amended. Members are kindly requested to indicate% ^ Ku Wu, Yi &quot; The film contains the d number of the order of the French method, the order of the letter 顺, the number of the first order from 8 to the number of less than 1 f. Note to the sequence of the sequence, it is beneficial to the special, please The sequence of the application process is as follows: the number of characters in the reading, a note, and a small note. If you want to enter the data entry area, the person who is doing the operation should produce the confession word in the following steps. Stored in words and phrases. Zhongzhongkuku materials and materials should be in the Yuci word word word language film 1cd &quot; 0 this fixed deposit decision deposit Θ mentioned Fan Li special application for the group position should be stored storage should be Checking the check I I. Therefore, the check of the square step words of the stored sentence is the same as the one: one phase and one small) 1¾ Jinzi brackets are more grouped. # 二 青 包 i Change the position t, 1 word from the beginning of the word, and the first word is the first word of the word. So the Fanli group _ the inventory material reserves are changed in this; according to the law, the procedure of the formula is changed. : Measure a few lines into the word I ----------- install · --------- order * -------- (Please read the precautions on the back before filling (In this page), you can install one package. Enter the 12-segment printing of the Intellectual Property Bureau employee consumer cooperative of the Ministry of Economic Affairs to print the split-sound reading, including the order, and then enter the phonetic system. Note the line. For notes, you can use the set to read the group; one mother to the group is one note to the input and the subcontracting area is included. So the sound is bought by the month -ηα Out of the application zone, 9 paper sizes are applicable to the Chinese National Standard (CNS) A4 (210 X 297 g) 527555 A8 B8 C8 D8. The scope of the patent application is to install the sub-voice sound. The group is in one application. Few back to the library and the material should be taken from the library to the reading library. The material system is subject to inspection and will be annoyed. Members are asked to indicate whether the amendments to the St / January amendments have been approved to modify the wording. The display of each display is displayed to show that the system should be used to check the response and to choose and identify. The marchers should exercise the confession with a word of 1 as little as one line. Including the check code package, the code is more suitable for reading in the sound, and the position, no, and advance; N one one is installed as ¾ points exchange 2 Unwrapping 11th tone sound reading range This group is conducive to a professional should be less Please go back to apply for the code and change the .3 code to 1 digit. The system 2 is compared with the simplified one mentioned in the 4th. If it is included, the system is checked. The system and the system stone The second transfer of the entry to the second paragraph should be conducive to the response and please, and then, the simplification and coding of the coded pieces, the simplification of the bar code, the decisive word and the decisive word. Take the reading code of the opposite reading than the installation and display the system code. The code is included in the phonetic transcription and the reading coupling. The module can be used to expand the word and include the information. It should be more coupled with the Chinese coupling. The pair of items is 2 to 11. The simplified formula should be helpful. For example, the materials in the recapitalization video should be added to the sequence language gc. You can use your own to use «/ 〇 library materials (please read the back Please fill in this page before refilling) ▼ Installation • m HI —I In nn vm 1 t ft mm In n # Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economy 20 This paper size applies to China National Standard (CNS) A4 (210 X 297g t)
TW88122058A 1999-12-15 1999-12-15 Intelligent phonetic input system and method therefor TW527555B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW88122058A TW527555B (en) 1999-12-15 1999-12-15 Intelligent phonetic input system and method therefor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW88122058A TW527555B (en) 1999-12-15 1999-12-15 Intelligent phonetic input system and method therefor

Publications (1)

Publication Number Publication Date
TW527555B true TW527555B (en) 2003-04-11

Family

ID=28787474

Family Applications (1)

Application Number Title Priority Date Filing Date
TW88122058A TW527555B (en) 1999-12-15 1999-12-15 Intelligent phonetic input system and method therefor

Country Status (1)

Country Link
TW (1) TW527555B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI425385B (en) * 2010-11-17 2014-02-01 Insyde Software Corp Small Chinese intelligence input method, computer-readable recording media and computer program products
TWI496031B (en) * 2013-01-08 2015-08-11
TWI689829B (en) * 2017-02-17 2020-04-01 李建文 Concentrated fast Pinyin input method and system

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI425385B (en) * 2010-11-17 2014-02-01 Insyde Software Corp Small Chinese intelligence input method, computer-readable recording media and computer program products
TWI496031B (en) * 2013-01-08 2015-08-11
TWI689829B (en) * 2017-02-17 2020-04-01 李建文 Concentrated fast Pinyin input method and system

Similar Documents

Publication Publication Date Title
TW411421B (en) System and method for processing chinese language text
WO2004027752A1 (en) Method and apparatus to facilitate correlating symbols to sounds
JP2008148322A (en) Method for processing character encoding, and system
Tran et al. Effective attention-based neural architectures for sentence compression with bidirectional long short-term memory
JP5323652B2 (en) Similar word determination method and system
Richmond et al. On generating Combilex pronunciations via morphological analysis
TW527555B (en) Intelligent phonetic input system and method therefor
Schmidt et al. Phonetic transcription standards for european names (onomastica).
Aida–Zade et al. The main principles of text-to-speech synthesis system
JP2000330996A (en) Pronouncing electronic dictionary
JP2010117529A (en) Device, method and program for generating voice reading sentence
JP4924148B2 (en) Pronunciation learning support device and pronunciation learning support program
JP2009199434A (en) Alphabetical character string/japanese pronunciation conversion apparatus and alphabetical character string/japanese pronunciation conversion program
JPS58123129A (en) Converting device of japanese syllabary to chinese character
JPH08339376A (en) Foreign language retrieving device and information retrieving system
Tsay Construction and automatization of a Minnan child speech corpus with some research findings
WO2024124697A1 (en) Speech recognition method, apparatus and device, and storage medium
Wickramasinghe et al. Practical issues in the development of TTS and SR for the Sinhala language
JPH0410052A (en) Preparing retrieving device for electronic dictionary of foreign language
Matogawa et al. Japanese Rule-based Grapheme-to-phoneme Conversion System and Multilingual Named Entity Dataset with International Phonetic Alphabet
Francis Annals of Vietnam: The Preservation of a Literary Heritage
TW444161B (en) Method and system providing similar code of character option
Fang A Prototype Text Analyzer for Mandarin Chinese TTS system
JPS61177573A (en) Forming device of japanese document
Hardegree Numeration systems

Legal Events

Date Code Title Description
GD4A Issue of patent certificate for granted invention patent
MM4A Annulment or lapse of patent due to non-payment of fees