514824 A7 五、發明説明（1 ) 發明部份 本發明係有關輸入裝置，用以配接電子裝置。更明確 言之’本發明係有關電腦鍵盤，語音及言語辨識系統，及 電子裝置之控制系統。 發明背景 ~ 先行技藝中知道語音偵測能力可與鍵盤整合。例如， 發給 Whelpley 之美專利 5 · 659 . 665 號（Whelpley 之、6 6 5專利）發表一種外部裝置，插入於鍵盤及電腦 間之資料電纜處。該外部裝置加語音辨識能力於電腦中， 變換語音信號爲鍵擊資料發送至電腦之鍵盤埠，同時鍵盤 本身所產生之普通信號可如常通過該外部裝置至電腦。 Whelpley之、6 6 5專利所發表之語音辨識系統·爲自含式 ，因爲語音輸入及語音辨識硬體限制於外部裝置中，語音 輸入信號之處理亦然。除與鍵資料相等者外，語音命令絕 不發送至電腦、且不消耗電腦處理時間來執行語音辨識功 能。514824 A7 V. Description of the invention (1) Inventive part The present invention relates to an input device for connecting an electronic device. More specifically, the present invention relates to a computer keyboard, a speech and speech recognition system, and a control system for electronic devices. BACKGROUND OF THE INVENTION ~ Advance technology knows that voice detection capabilities can be integrated with keyboards. For example, US Patent No. 5.659.665 issued to Whelpley (patented by Whelpley, patent No. 665) discloses an external device that is inserted into a data cable between a keyboard and a computer. The external device adds speech recognition capability to the computer, converts the voice signal into keystroke data and sends it to the keyboard port of the computer, and at the same time, the ordinary signals generated by the keyboard itself can pass through the external device to the computer as usual. The voice recognition system published by Whelpley's patent, 6 6.5 is self-contained, because the voice input and voice recognition hardware are limited to external devices, so is the processing of voice input signals. Except for those equivalent to key data, voice commands are never sent to a computer and do not consume computer processing time to perform speech recognition functions.
Whelpley之、6 6 5專利亦發表一實施例，其中，鍵 盤機匣內包含語音辨識裝置。與外部實施例同樣，裝置用 以變換語音信號爲鍵擊資料，並插入該資料於來自鍵盤之 資料流中。W.helpley之~ 6 6 5專利中所發表之組合有效 掩飾語音辨識功能，俾使電腦透通。故此，語音信號絕不 在鍵盤電纜上發送。 僅'' 模仿〃鍵盤操作之任何先行技藝裝置，諸如 —---------- (請先閲讀背面之注意事項再填寫本頁) 訂 經濟部智慧財產局員工消費合作社印製Whelpley's patent No. 665 also discloses an embodiment in which the keyboard case contains a voice recognition device. As in the external embodiment, the device converts the voice signal into keystroke data and inserts the data into the data stream from the keyboard. W.helpley's ~ 6 6 5 patent combination effectively masks the voice recognition function and makes the computer transparent. Therefore, voice signals are never sent on the keyboard cable. Only '' imitates any prior art device operated by the keyboard, such as ----------- (Please read the notes on the back before filling this page) Order Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs
514824 A7 _____ B7____ 五、發明説明（2 )514824 A7 _____ B7____ 5. Description of the invention (2)
Whelpley之、6 6 5專利所發表之裝置有效限制功能於鍵 盤輸入及可由鍵擊組合執行之命令。此爲重大缺點’因爲 電腦無法對語音信號本身執行操作，而無重複之微音器/ .揚聲器組合直接連接至電腦，且不依賴辨識系統。 且知道語音辨識爲一演進之技術。現行實施可爲'' 語 音依賴〃者，其中，系統設計精確辨識特定使用者之語音 ，或爲 > 語音獨立〃者，其中，系統設計精確作用於任何 語音。此方面之進步已降低由個別發音，說話樣式，及轉 換言語爲鍵擊之重點之不同所引起之不精確性。由於任何 自含之語音辨識實施易受限於其內之硬體，故當技術演進 時，升級之費用高。 而且，諸如基本文字輸入等功能由於需要'' 訓練〃現 .語音辨識系統，以精確認識特定使用者之語音，·而大爲複 雜。發給Galvin之美專利5，8 7 4，9 3 9號（The device published by Whelpley, Patent No. 6 65 effectively restricts functions to keyboard input and commands that can be executed by keystroke combinations. This is a major disadvantage because the computer cannot perform operations on the voice signal itself, and the non-repeating microphone / speaker combination is directly connected to the computer and does not rely on the identification system. And know that speech recognition is an evolved technology. The current implementation can be a `` speech-dependent person '', in which the system design accurately recognizes the speech of a specific user, or a > speech independent person, in which the system design accurately acts on any speech. Improvements in this area have reduced the inaccuracies caused by differences in individual pronunciations, speaking styles, and the key focus of translating speech. As any self-contained speech recognition implementation is easily constrained by the hardware inside it, the upgrade cost is high as the technology evolves. Moreover, functions such as basic text input require “training to emerge.” Speech recognition systems to accurately recognize the voice of a particular user are greatly complicated. No. 5,8 7 4,9 3 9 issued to Galvin (
Galvm < 9 3 9專利〃）發表一種語音辨識鍵盤，具有一 顯示器，以便訓練。當口述語句由系統變換爲文字時，使 用者可在顯示器上檢查其正確性，且如此，經過時間提高 系統之精確性。 與 Whelpley 之 ~ 6 6 5 專利同樣，Galnn、9 3 9 專 利中所發表之語音辨識系統爲自含式，且並不發送語音資 料至電腦。結果，訓練程序需受限於鍵盤本地之介面，如 所發表，此可稀少或昂貴，且當然，有設置鍵盤顯示器， 以及在所連接之電腦上顯示之問題。 雖Whelpley之~ 6 6 5專利及Galvm ' 9 3 9專利所 本紙張尺度適用中國國家標準（CNS ) A4規格（21〇Χ29<7公釐） (請先閱讀背面之注意事項再填寫本頁) -裝·Galvm < 9 3 9 patent 〃) published a speech recognition keyboard with a display for training. When the spoken sentence is changed from the system to the text, the user can check the correctness on the display, and in this way, the accuracy of the system is improved over time. Like Whelpley's ~ 6.5 patent, the speech recognition system published in Galnn, 9 3.9 patents is self-contained and does not send voice data to the computer. As a result, the training program needs to be limited to the local interface of the keyboard. As published, this can be rare or expensive, and of course, there are problems with setting up a keyboard display and displaying it on a connected computer. Although Whelpley's ~ 6 5 5 patent and Galvm '9 3 9 patent, the paper size of this paper applies the Chinese National Standard (CNS) A4 specification (21〇 × 29 < 7 mm) (Please read the precautions on the back before filling this page) -Loaded ·
、1T 經濟部智慧財產局員工消費合作社印製 514824 A7 ________B7_ 五、發明説明（3 ) 發表之自含方法欲由限制言語之處理於與電腦系統隔離之 外部處理器中，以提高系統效率，且可適用於較舊之電腦 系統，但其彈性不足，無法實際使用於現行電腦系統及現 有之高數元率介面上，且並不輔助其他語音有關之技藝之 使用。 發明槪要 需要低成本之裝置，能同時發送語音及鍵盤信號經同 一介面至電腦。且需要一種發音辨識系統，此可模仿鍵盤 信號，以及支持其他語音處理功能，諸如數位電話及遙控 裝置。需要非局限於鍵盤之環境中之此等功能。亦需要一 種發音辨識系統，此可自鍵盤內作用，或與電腦系統結合 ，視應用而定。 本發明之目的在整合來自電腦鍵盤之鍵擊資料及其他 周邊介面，諸如微音器或電話手機及發音辨識系統之語音 資料，用於文字輸入，控制電子裝置，及進出網際網路。 本發明之一實施例在提供一輸入裝置，諸如普通電腦 鍵盤，此適於與語音信號之輸入共用其介面控制器，並使 信號可在攜帶鍵擊資料之同一電纜上發送至電腦系統。 由使用充分高數元率之鍵盤介面控制器，諸如通用串 列匯流排（'' U S B 〃 ）控制器，達成在單條電纜上傳輸 鍵盤及語音信號二者至電腦，及連接至控制器之一語音處 理器變換來自例如微音器之類比語音信號爲數位。如此， U S B鍵盤變爲語音信號進出電腦之導管。如此，可輔助 本紙張尺度適用中.國國家標準（CNS ) A4規格（210X297公釐） " -6 - (請先閱讀背面之注意事項再填寫本頁) 裝 訂 經濟部智慧財產局員工消費合作社印製 514824 A7 _B7_ 五、發明説明（4 ) 或取代現有之周邊裝置，諸如電腦系統中現用以處理聲音 之語音及聲音卡。 (請先閱讀背面之注意事項再填寫本頁) 諸如揚聲器之語音輸出至語音處理器之連接可作雙工 .語音傳輸，可與電信應用相容。記憶器之加裝於語音處理 器中延伸輸入裝置之能力至語音發訊，語音印跡識別，及 語音辨識。 本發明之目的由設置電腦系統之一輸入裝置達成，包 含一介面控制器用以提供信號用於電腦系統中；及一鍵擊 輸入裝置連接至該介面控制器；及一語音輸入裝置連接至 該介面控制器，鍵擊輸入裝置及語音輸入裝置共用介面控 制器，以發送鍵盤及語音資料至電腦。 本發明之另一實施例爲一語音辨識系統，其中使用一 經濟部智慧財產局員工消費合作社印製 _單個輸入裝置來接收語音及鍵盤資訊，電腦使用語音辨識 技術處理語音輸入，俾作爲文字輸出，或由執行對應之操 作指令。如上述，輸入裝置可爲一鍵盤，諸如一 U S B鍵 盤，及一語音處理器可置於其內，以接受類比語音輸入。 可加裝記憶器於輸入裝置內或電腦中，以方便語音印跡辨 識。可使用輸入裝置及電腦間之信號之譯密，以防止窺察 介面。語音辨認亦可由輸入裝置與電腦平行達成，以分配 計算負荷於系統上，或提高語音辨識功能之精確性。 此中並提供一語音辨識系統，包含：一電腦系統具有 一發話辨識裝置連接至輸入處理裝置及輸出裝置；及一輸 入裝置用以發送語音及鍵擊信號，具有鍵擊輸入裝置及語 音輸入裝置，連接並共用至少一介面控制器，連接至電腦 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） ' ~ 514824 A7 ___B7 五、發明説明（5 ) 之輸入處理裝置；其中發話辨裝置處理語音信號，並產生 輸出至輸出裝置。 在另一實施例，語音辨識裝於一鍵盤中，具有一信號 .裝置能發送資料至電腦，以及輸入至一或更多之外部裝置 。此實施例之一典型處理爲使用U S B技術配接電腦，及 紅外線（I R )可用以控制電視，或射頻（R F )傳輸於 應用，如無線電話。執行言語辨識所需之組成件包含於鍵 盤中，以變換言語爲文字，經鍵盤介面傳輸至電腦，並提 供語音操作之遙控給任何可相容之電子裝置。 且提供電腦系統之一語音辨識鍵盤，用以產生鍵擊輸 入及操作指令、，用以控制至少一電子裝置，包含：一信號 裝置用以發送可能爲二式之信號，第一式適於傳輸至電腦 系統，及第二式適於傳輸至至少一電子裝置·，及·多個鍵連 接至信號裝置，用以提供鍵擊輸入至信號裝置，俾傳輸通 過其中；及一語音輸入裝置；及一語音辨識裝置連接至語 音輸入裝置，以變換口述語句爲操作指令，俾傳輸通過信 號裝置。 需要至少一語音輸入裝置，以提供語音輸入至鍵盤及 語音處理電路，以執行信號處理及語音辨識。如在前實施 例中，加記憶器，譯密電路，及語音輸出裝置於鍵盤，俾 可作語音印跡.比較及雙路語音通訊，適於確保數位電話及 語音資料輸入。 在另一實施例，設置一輸入裝置，諸如鍵盤，當連接 至構造可進出網際網路之電腦時，於收到特定之口述命令 本紙張尺度適用中國國家標準（CNS ) A4規格（210 X 297公釐） (請先閲讀背面之注意事項再填寫本頁) 裝· 、11 經濟部智慧財產局員工消費合作社印製 -8- 514824 A7 _B7_ 五、發明説明（6 ) 時，此發動進出網際網路。 提供一語音辨識輸入裝置，用以提供自構造連接於網 際網路之電腦由語音進出該網際網路，包含一語音輸入裝 .置；及一語音辨識裝置用以辨別口述命令，命令之至少之 一訂定發動網際網路進出；及一信號裝置用以觸發該命令 之執行。 本發明並構想一種方法，用以提供鍵盤及語音輸入經 一單個介面至電腦系統，變換口述語句爲文字，自連接至 網際網路之電腦進出該網際網路，及由口述命令遙控至少 一電子裝置。 自以下說明並參考附圖，可明瞭本發明之以上及其他 特色，層面，及優點，在附圖中，相同之參考編號指相同 .之元件。 附圖簡述 本發明由實例圖解，且並非限制於附圖中，在附圖中 ，相同之參考編號標示相當之部份，且其中： 圖1爲方塊圖，顯示本發明之輸入裝置； 圖2爲方塊圖，顯示依本發明整合輸入裝置於U S B 鍵盤中之一實施例； 圖3爲方塊圖，顯示連接至電腦系統之本發明之具有 語音辨識裝置之輸入裝置之一實施例； 圖4爲方塊圖，顯示連接至電腦系統之本發明之輸入 裝置之一實施例，電腦另連接至外部服務裝置； 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） " -9 - (請先閱讀背面之注意事項再填寫本頁) -裝· 訂 經濟部智慧財產局員工消費合作社印製 514824 Μ ___Β7 五、發明説明（7 ) 圖5爲方塊圖，顯示本發明之一實施例，其中，整合 之輸入裝置爲具有遙控能力之一語音辨識鍵盤；及 (請先閱讀背面之注意事項再填寫本頁) 圖6爲方塊圖，顯示連接至可進出網際網路之電腦之 .本發明之一實施例。 主要元件對照表個數 1〇：言語輸入裝置 1 1 :機箱 1 2 :介面控制器 1 3 :鍵盤輸入 1 4 :語音處理器 1 6 :記憶器 1 7 :語音輸入 1 8 :語音輸出 1 9 :信號譯密裝置 經濟部智慧財產局員工消費合作社印製 2 0 :電腦 2 1 :輸入處理器 2 2 :目吾首副系統 2 3 :輸出裝置 2 6 ·· I /〇控制器 3 0 :電子裝置 較佳實施例之說明 現參考圖1 ，其中顯示本發明之言語輸入裝置之較佳 本紙張尺度適用中國國家標準（CNS ) Α4規格（210Χ297公釐） -10 - 514824 A7 _____B7_ 五、發明説明（8 ) 實施例之電方塊圖，整個由數字1 〇標示。槪要顯示言語 輸入裝置1 〇之一機箱1 1。一電腦系統整個由數字2〇 標示。電腦系統2 0具有一輸入處理器2 1經連接，俾與 .語音輸入裝置1 〇之介面控制器1 2雙向通訊。介面控制 器1 2經連接，以接收來自鍵擊輸入1 3及語音輸入1 7 之信號。 一語音處理器1 4連接於語音輸入1 7及介面控制器 1 2之間。語音處理器1 4及介面控制器1 2間有一雙向 連接存在。介面控制器1 2及語音處理器1 4可合倂成單 個晶片。功能應相同。語音輸出i 8及記憶器1 6連接至 語音處理器1 4。應明瞭雖語音輸入1 7及語音輸出1 8 在圖1顯示在機箱1 1內，但此等可置於機箱外，而不影 響本發明之功能及精神。 在操作中，自語音輸入1 7接收之信號由語音處理器 1 4數位化，及語音資料發送至介面控制器1 2。介面·控 制器1 2合倂該語音資料及自鍵擊輸入1 3所扱收之鍵擊 資料’並發送此等作爲單個資料信號至電腦系統2 0之輸 入處理器2 1。在裝有U S B之本裝置中，鍵盤及微音器 使用二不同之端點作爲一複合ϋ S B裝置，故鍵盤資料及 語音資料無需同時。語音資料可爲等時或非等時。如該資 料爲等時’則.此在U S Β極流排上具有較所有其他型式之 資料爲高之優先。射頻資料亦可爲非等時，並與另一 U S Β.競爭。不管使用何介面方法，語音資料及鍵擊資料 可同時發送，或一次發送一或另一。語資料及鍵擊資料發 本紙適用中國國家標準（CNS： ) Α4規格（210X297公嫠） ~ -- (請先閲讀背面之注意事項再填寫本頁) 裝· 訂 經濟部智慧財產局員工消費合作社印製 -11 - 514824 A 7 B7 五、發明説明（Θ ) 送經單條電纜至電腦系統。 (請先閱讀背面之注意事項再填寫本頁) 同樣，含有由介面控制器1 2自電腦2 0接收之語音 資料之信號發送至語音處理器，以變換爲適用於語音輸出 1 8之類比信號。在簡單之情形，語音處理器用作輸入之 類比至數位變換器（A D C )，及用作語音資料輸出之數 位至類比變換器。 可處理自語音輸入1 7所接收之信號，用於任何型式 之發音辨識，諸如說話辨識，此變換人類說話爲文字，及 語音辨識，此處理人類說話爲命令或執行確認。 記憶器1 6可包含一語音印跡樣品或語音辨識資料， 使語音處理器1 4能執行來自語音輸入1 7之語音信號之 語音印跡比較，或對其執行語音辨識功能。 圖2爲與圖1相似之本發明之一電功能方塊圖，其中 ，言語輸入裝置以一 U S B鍵盤爲基礎。u S B鍵盤之標 準組成件安排成典型形態，包含介面控制器1 2連接至鍵 擊輸入13及信號LED1 5，並具有一 USB輸出。 經濟部智慧財產局員工消費合作社印製 有數類U S B存在。一 U S B鍵盤本身爲人類介面裝 ft ( Η I D )類。其他類包含聲音裝置，通訊裝置，顯示 裝置，及大量儲存裝置類。聲音裝置類定義爲U S Β微音 器之一裝置說明器。使用U S Β聲音裝置類說明器爲支持 U S Β微音器，之一標準方式，且需要每秒全速1 2百萬數 元U S Β介面晶片，此支持等時轉移。此一晶片較之低速 晶片昂貴’但確保與微音器之較大範圍之U S Β系統驅動 器相容。然而，由於本發明構想等時及非等時資料傳輸， 本紙張尺度適用中國國家標準（CNS ) Α4規格（210X297公釐） -12- 514824 A7 B7 經濟部智慧財產局員工消費合作社印製 五、發明説明（10) 故任何介面可取代此等U s B裝置類，而不改變其基本功 tb 。 本發明加一'語苜處理器1 4於U S B鍵盤中，連接至 .介面控制器1 2，及分別代表微音器及揚聲器之語音輸入 1 7及語音輸出1 8。語音輸入1 7可爲一板上微音器（ 即與鍵盤整合），或插入於插口連接器中之一微音器。記 憶器1 6亦連接至語音處理器1 4。 言語輸入裝置1 0之操作與圖1所述相似，由介面控 制器1 2合倂自鍵擊輸入1 3所接收之資料及自語音輸入 1 7所接收之語音資料，並由語音處理器1 4數位化。數 位信號進入介面控制器1 2中，此用作U S B處理器，此 訊包化該等信號。結果信號發送於U S B電纜1 9上，作 爲標準U S B技術之語音訊包，可與具有相似U S B介面 之任何電腦系統相容。訊包之一可能格式爲1 6數元8 K Η z P CM之U S B微音聲頻資料格式；然而，有許多其 他可能之聲頻格式。通過U S B電纜1 9進來之語音信號 由介面控制器1 2轉送至語音處理器1 4，俾經由語音輸 出1 8輸出。語音處理器1 4及介面控制器1 2可合倂成 單個晶片，而不影響功能。L E D 1 5普通用作鍵盤指示 燈，諸如數鎖或大寫鎖，但亦可用以指示語音輸入裝置 1 0之其他方面之指示功能。 在此構造中，語音輸入裝置1 0作用恰如一標準 U S B.鍵盤，且亦如語音信號之雙路導管，而不加額外之 介面於U S B順應電腦。而且，語音輸入1 7及語音輸出 (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（210X29*7公釐） -13- 514824 A7 _ B7 五、發明説明（11) (請先閲讀背面之注意事項再填寫本頁) 1 8可合倂爲一接收單位’用作電話手機。加裝記憶器 1 6於語音處理器1 4使系統更能由比較進來之信號及記 憶器1 6中所儲存之信號，檢查使用者身份，或限制語音 輸入裝置1 0之進出於特定核准之使用者。在本構造中亦 可加裝任何發音辨識功能於語音處理器1 4中，同時保持 單個輸入裝置之優雅形狀。 現參考圖3 ，方塊圖顯示發音辨識系統之較佳實施例 ’使用輸入裝置1 0 ’其中’電腦系統2 0具有一輸入處 理器2 1連接至語音副系統裝置2 2 ，此連接至一輸出裝 置2 3。輸入裝置1 0具有與圖1相同之基本組成件，並 同樣操作，以提供鍵擊及語音資料給電腦系統2 〇 ,經一 單個介面至輸入處理器2 1。語音資料由輸入處理器2 1 隔離，俾傳輸至語音副系統裝置2 2。依據此實施例，語 音副系統裝置可包含電腦系統之自然處理裝置，執行一組 指令形態之軟體，或可由特殊化之硬體或二者之合倂構成 〇 經濟部智慧財產局員工消費合作社印製 輸出裝置2 3可提供語音副系統裝置2 2之直接輸出 ’諸如口述語句變換爲文字顯示於監視器上，或可爲較間 接之輸出，諸如語音聽寫至語句處理器，並輸出至印表機 。與前實施例同樣，輸入裝罝1 〇之介控制器1 2可修改 ，俾依據U S β技術產生信號。同樣，話音處理器1 4可 連接於語音輸入1 7及介面控制器1 2之間，以提供比至 數位變換。可設置一語音輸出1 8，在此情形，語音處理 器1 4變換經由介面控制器1 2所接收之數位語音信號爲 本紙張尺度適用中國國家標準（CNS ) Α4規格（210X297公釐） -14- 514824 A7 B7 五、發明説明（12) 適合輸出至揚聲器之類比信號。 與在前實施例中同樣，語音處理器1 4之功能無需限 於信號模式變換。記憶器1 7之加裝可在輸入裝置1 0本 .地處理語音，諸如語音印跡比較及言語辨識。語音印跡比 較可用以比較進來之語音信號及記憶器1 6中所記錄之語 音印跡。經由語音處理器1 4之本地發音辨識可與語音副 系統裝置2 2聯合使用，在此情形，輸入裝置1 〇用作電 腦系統2 0之協同處理器。 顯示於輸出裝置2 3上之使用者介面可爲視窗基礎者 ，且可由拉下選單驅動，其中顯示語音命令。故此，使用 者無需記憶由該系統所辨識之命令。此方法可使發音辨識 系統對使用者更爲可親。當然，可使用任何顯示介面，或 可量身訂製一介面，以執行功能，而不改變本發明之功能 〇 圖4顯不圖3之發苜辨識系統之一修改型實施例，其 中，具有槪要表示之機箱1 1之輸入裝置1 0具有介面控 制器1 2經連接而接收來自鍵擊輸入1 3之鍵擊資料，且 經連接而接收來自語音處理器之語音輸入。介面控制器 1 2使數位語音信號訊包化，以發送其經電纜至電腦系統 2 0。語音輸入1 7至語音輸出1 8分別發送及接收自語 音處理器1 4 .通過介面控制器1 2經同一電纜而至電腦系 統2 0之類比語音資料。與前實施例同樣，記憶器1 6由 語音處理器1 4使用，以儲存資料，供語音印跡比較及語 音辨識之用。 本&張尺度適用中國國家標準（CNS ) A4規格（210X297公釐）—" ~ (請先閱讀背面之注意事項再填寫本頁) •裝·1T printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 514824 A7 ________B7_ V. The self-contained method published in the description of the invention (3) is intended to limit the processing of speech to an external processor isolated from the computer system to improve system efficiency, and It can be applied to older computer systems, but it is not flexible enough to be used in current computer systems and existing high-bit-rate interfaces, and does not assist the use of other speech-related techniques. The invention requires a low-cost device that can simultaneously send voice and keyboard signals to a computer via the same interface. There is also a need for a pronunciation recognition system that mimics keyboard signals and supports other voice processing functions, such as digital phones and remote controls. There is a need for these functions in environments other than keyboards. A pronunciation recognition system is also needed, which can be used from the keyboard or combined with a computer system, depending on the application. The purpose of the present invention is to integrate keystroke data from a computer keyboard and other peripheral interfaces, such as voice data from a microphone or a telephone handset and a pronunciation recognition system, for text input, control of electronic devices, and access to the Internet. One embodiment of the present invention provides an input device, such as a common computer keyboard, which is suitable for sharing its interface controller with the input of voice signals, and enables the signal to be sent to the computer system on the same cable carrying the keystroke data. By using a keyboard controller with a sufficiently high data rate, such as a universal serial bus ('' USB 〃) controller, it is possible to transmit both keyboard and voice signals to a computer on a single cable, and to connect to one of the controllers The speech processor converts analog speech signals from, for example, a microphone into digital. In this way, the USB keyboard becomes a conduit for voice signals to and from the computer. In this way, it can assist in the application of this paper standard. National Standard (CNS) A4 specification (210X297 mm) " -6-(Please read the precautions on the back before filling this page) Binding of the Intellectual Property Bureau's Employee Consumption Cooperatives Printed 514824 A7 _B7_ 5. Description of the Invention (4) or replace the existing peripheral devices, such as the voice and sound card currently used to process sound in computer systems. (Please read the notes on the back before filling this page) The connection between the voice output of the speaker and the voice processor can be duplex. The voice transmission is compatible with telecommunications applications. The memory is installed in the speech processor to extend the input device's capabilities to speech transmission, speech imprint recognition, and speech recognition. The purpose of the present invention is achieved by setting an input device of a computer system, including an interface controller for providing signals for use in the computer system; and a keystroke input device connected to the interface controller; and a voice input device connected to the interface The controller, key input device and voice input device share the interface controller to send keyboard and voice data to the computer. Another embodiment of the present invention is a speech recognition system in which a consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs prints a single input device to receive speech and keyboard information, and the computer uses speech recognition technology to process the speech input and use it as text output , Or by executing the corresponding operation instruction. As described above, the input device may be a keyboard, such as a USB keyboard, and a voice processor may be placed therein to accept analog voice input. A memory can be installed in the input device or in the computer to facilitate the recognition of the voice print. Interpretation of signals between input devices and computers can be used to prevent spying on the interface. Speech recognition can also be achieved in parallel with the input device and the computer to distribute the computational load on the system or improve the accuracy of the speech recognition function. A speech recognition system is also provided, including: a computer system having a speech recognition device connected to an input processing device and an output device; and an input device for sending voice and keystroke signals, having a keystroke input device and a voice input device , Connect and share at least one interface controller, connect to the computer paper size applicable to Chinese National Standard (CNS) A4 specification (210X297 mm) '~ 514824 A7 ___B7 V. Input processing device of invention description (5); The device processes the speech signal and generates output to an output device. In another embodiment, the speech recognition is installed in a keyboard and has a signal. The device can send data to the computer and input to one or more external devices. A typical process of this embodiment is to use USB technology to connect to a computer, and infrared (IR) can be used to control television, or radio frequency (RF) is transmitted to applications, such as wireless phones. The components required to perform speech recognition are contained in a keyboard, which converts speech into text, transmits it to a computer via a keyboard interface, and provides remote control for voice operations to any compatible electronic device. A speech recognition keyboard for a computer system is provided to generate keystroke inputs and operation instructions to control at least one electronic device, including: a signaling device for sending signals that may be of two types, the first type being suitable for transmission To a computer system, and the second type is adapted to be transmitted to at least one electronic device, and a plurality of keys are connected to the signal device for providing keystroke input to the signal device through which transmission is transmitted; and a voice input device; and A speech recognition device is connected to the speech input device, and converts the spoken sentence as an operation instruction, and transmits it through the signal device. At least one voice input device is required to provide voice input to the keyboard and voice processing circuit to perform signal processing and voice recognition. For example, in the previous embodiment, a memory, a transcoding circuit, and a voice output device are added to the keyboard, which can be used for voice imprinting. Comparison and two-way voice communication are suitable for ensuring digital telephone and voice data input. In another embodiment, an input device, such as a keyboard, is provided. When connected to a computer that is configured to access the Internet, a specific dictation command is received. This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297). (Mm) (Please read the precautions on the back before filling this page). · 11, Printed by the Consumers' Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs -8- 514824 A7 _B7_ 5. When the invention description (6), this launches in and out of the Internet road. A speech recognition input device is provided for providing a computer connected to the Internet from and to the Internet to and from the Internet, including a speech input device; and a speech recognition device for distinguishing spoken commands, at least one of the commands An order to initiate Internet access; and a signaling device to trigger the execution of the command. The present invention also contemplates a method for providing a keyboard and voice input to a computer system via a single interface, converting a spoken sentence into text, entering and exiting the Internet from a computer connected to the Internet, and remotely controlling at least one electronic by a spoken command Device. The above and other features, aspects, and advantages of the present invention will be apparent from the following description and reference to the accompanying drawings. In the drawings, the same reference numerals refer to the same elements. Brief Description of the Drawings The present invention is illustrated by examples and is not limited to the drawings. In the drawings, the same reference numerals indicate corresponding parts, and among them: Figure 1 is a block diagram showing an input device of the present invention; 2 is a block diagram showing an embodiment of an integrated input device in a USB keyboard according to the present invention; FIG. 3 is a block diagram showing an embodiment of an input device with a voice recognition device of the present invention connected to a computer system; FIG. 4 It is a block diagram showing an embodiment of the input device of the present invention connected to a computer system, and the computer is also connected to an external service device; This paper size applies to the Chinese National Standard (CNS) A4 specification (210X297 mm) " -9- (Please read the precautions on the back before filling out this page)-Binding and printing printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 514824 M ___ Β7 5. Description of the invention (7) Figure 5 is a block diagram showing an embodiment of the present invention Among them, the integrated input device is a voice recognition keyboard with remote control capability; and (please read the precautions on the back before filling this page) Figure 6 is a block diagram showing the connection Allowing access to the computer of the Internet. One embodiment of the present invention. Number of comparison tables of main components 10: Speech input device 1 1: Chassis 1 2: Interface controller 1 3: Keyboard input 1 4: Voice processor 1 6: Memory 1 7: Voice input 1 8: Voice output 1 9 : Signal translation device printed by the Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economics 2 0: Computer 2 1: Input Processor 2 2: Megou Chief System 2 3: Output Device 2 6 ·· I / 〇 Controller 3 0: The description of the preferred embodiment of the electronic device will now be referred to FIG. 1, which shows the preferred speech input device of the present invention. The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 × 297 mm) -10-514824 A7 _____B7_ V. Invention Explanation (8) The electrical block diagram of the embodiment is entirely indicated by the number 10.槪 To display one of the speech input devices 1 0 chassis 1 1. A computer system is indicated by the number 20 as a whole. The computer system 20 has an input processor 21, which is in communication with the interface controller 12 of the voice input device 10 through two-way communication. The interface controller 12 is connected to receive signals from the key input 13 and the voice input 17. A voice processor 14 is connected between the voice input 17 and the interface controller 12. A two-way connection exists between the voice processor 14 and the interface controller 12. The interface controller 12 and the voice processor 14 can be combined into a single chip. The functions should be the same. The voice output i 8 and the memory 16 are connected to the voice processor 14. It should be understood that although the voice input 17 and the voice output 18 are shown in the case 11 in FIG. 1, these can be placed outside the case without affecting the function and spirit of the present invention. In operation, the signal received from the voice input 17 is digitized by the voice processor 14 and the voice data is sent to the interface controller 12. The interface controller 12 combines the voice data and the keystroke data received from the keystroke input 13 and sends these as a single data signal to the input processor 21 of the computer system 20. In this device equipped with USB, the keyboard and microphone use two different endpoints as a composite ϋSB device, so the keyboard data and voice data need not be simultaneously. Voice data can be isochronous or non-isochronous. If the data is isochronous' then this has priority over all other types of data on the U S Β pole bus. Radio frequency data may also be non-isochronous and compete with another U S B. No matter which interface method is used, voice data and keystroke data can be sent simultaneously, or one or the other at a time. Language documents and keystroke documents are issued in accordance with Chinese National Standards (CNS :) Α4 specifications (210X297 gong) ~-(Please read the precautions on the back before filling out this page) Printing -11-514824 A 7 B7 5. Description of the Invention (Θ) Send a single cable to the computer system. (Please read the precautions on the back before filling this page) Similarly, the signal containing the voice data received by the interface controller 12 from the computer 20 is sent to the voice processor to be converted into an analog signal suitable for voice output 1 8 . In the simple case, a speech processor is used as an analog-to-digital converter (A D C) for input and a digital-to-analog converter for speech data output. It can process signals received from voice input 17 for any type of pronunciation recognition, such as speech recognition, which transforms human speech into text, and speech recognition, which processes human speech as a command or performs confirmation. The memory 16 may include a voice imprint sample or voice recognition data, so that the voice processor 14 can perform a voice imprint comparison of a voice signal from the voice input 17 or perform a voice recognition function on it. FIG. 2 is a block diagram of an electrical function of the present invention similar to FIG. 1, wherein the speech input device is based on a USB keyboard. The standard components of the USB keyboard are arranged in a typical form, including the interface controller 12 connected to the key input 13 and the signal LED 15, and has a USB output. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs, several types of USB exist. A U S B keyboard itself is a ft (Η ID) type of human interface. Other categories include sound devices, communication devices, display devices, and mass storage devices. The sound device class is defined as one of the device explainers of the USB microphone. The use of a USB sound device class specifier is a standard way to support a USB sound device, and requires a full-speed U.S. 12 million U.S. dollars interface chip per second, which supports isochronous transfer. Such a chip is more expensive than a low speed chip 'but ensures compatibility with a larger range of USB system drivers for microphones. However, due to the concept of isochronous and non-isochronous data transmission of the present invention, the paper size is applicable to the Chinese National Standard (CNS) A4 specification (210X297 mm) -12- 514824 A7 B7 Printed by the Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs Description of the invention (10) Therefore, any interface can replace these U s B device classes without changing its basic work tb. In the present invention, a language processor 14 is connected to a USB keyboard, and is connected to the interface controller 12 and voice input 17 and voice output 18 respectively representing a microphone and a speaker. The voice input 17 can be a microphone on the board (that is, integrated with the keyboard) or a microphone inserted into the socket connector. The memory 16 is also connected to the speech processor 14. The operation of the speech input device 10 is similar to that described in FIG. 1. The interface controller 1 2 combines the data received from the key input 1 3 and the speech data received from the speech input 17. The speech processor 1 4 digitization. The digital signals enter the interface controller 12, which is used as a USB processor, and this signal encapsulates these signals. The resulting signal is sent on the USB cable 19 as a voice packet of standard USB technology and is compatible with any computer system with a similar USB interface. One possible format of the packet is a 16-digit 8 K Η z P CM U S B microphonic audio data format; however, there are many other possible audio formats. The voice signal coming in via the USB cable 19 is transferred from the interface controller 12 to the voice processor 14 and then output via the voice output 18. The voice processor 14 and the interface controller 12 can be combined into a single chip without affecting the function. L E D 1 5 is generally used as a keyboard indicator, such as a number lock or a caps lock, but it can also be used to indicate other aspects of the voice input device 10's indicating function. In this configuration, the voice input device 10 functions just like a standard U S B. keyboard, and also acts as a two-way conduit for voice signals, without adding an additional interface to the U S B conforming to the computer. Moreover, voice input 17 and voice output (please read the notes on the back before filling this page) This paper size is applicable to China National Standard (CNS) A4 specification (210X29 * 7 mm) -13- 514824 A7 _ B7 V. Description of the Invention (11) (Please read the precautions on the back before filling out this page) 1 8 can be combined into a receiving unit 'used as a telephone handset. Adding memory 16 to voice processor 14 enables the system to compare the incoming signals and the signals stored in memory 16 to check the identity of the user or restrict the access of voice input device 10 to specific approvals. user. In this configuration, any pronunciation recognition function can be added to the speech processor 14 while maintaining the elegant shape of a single input device. Referring now to FIG. 3, a block diagram shows a preferred embodiment of the pronunciation recognition system 'using an input device 1 0', where 'computer system 2 0 has an input processor 2 1 connected to a voice subsystem system device 2 2, which is connected to an output Device 2 3. The input device 10 has the same basic components as those in FIG. 1 and operates in the same way to provide keystrokes and voice data to the computer system 2 0 through a single interface to the input processor 21. The voice data is isolated by the input processor 2 1 and transmitted to the voice sub-system device 2 2. According to this embodiment, the voice auxiliary system device may include a natural processing device of a computer system, software that executes a set of instructions, or may be composed of specialized hardware or a combination of both. The output device 23 can provide the direct output of the voice sub-system device 2 2 'such as converting spoken sentences to text displayed on a monitor, or it can be a more indirect output, such as speech dictation to a sentence processor and output to a print machine. As in the previous embodiment, the medium controller 12 of the input device 10 can be modified to generate a signal according to the U S β technology. Similarly, a voice processor 14 can be connected between the voice input 17 and the interface controller 12 to provide a ratio-to-digital conversion. A voice output 18 can be set. In this case, the voice processor 14 converts the digital voice signal received through the interface controller 12 to the paper size. The Chinese National Standard (CNS) Α4 specification (210X297 mm) -14 -514824 A7 B7 V. Description of the invention (12) Suitable for analog signal output to speakers. As in the previous embodiment, the function of the speech processor 14 need not be limited to signal mode conversion. The addition of the memory 17 can process speech at the input device 10, such as the comparison of speech imprints and speech recognition. The phonetic print comparison can be used to compare the incoming voice signal with the phonetic print recorded in the memory 16. The local pronunciation recognition through the speech processor 14 can be used in conjunction with the speech auxiliary system device 22. In this case, the input device 10 is used as a coprocessor of the computer system 20. The user interface displayed on the output device 23 can be a window-based user, and can be driven by a pull-down menu, in which voice commands are displayed. Therefore, the user does not need to memorize the commands recognized by the system. This method makes the pronunciation recognition system more accessible to users. Of course, any display interface can be used, or an interface can be customized to perform the function without changing the function of the present invention. Fig. 4 shows a modified embodiment of the hair recognition system of Fig. 3, which has槪 The input device 10 of the chassis 11 to be shown has an interface controller 12 that receives keystroke data from the keystroke input 13 through connection, and receives voice input from the voice processor via connection. The interface controller 1 2 packetizes the digital voice signal to send it to the computer system 20 via a cable. Voice input 17 to voice output 18 are sent and received from the voice processor 14 respectively. Analog voice data through the interface controller 12 to the computer system 20 via the same cable. As in the previous embodiment, the memory 16 is used by the speech processor 14 to store data for comparison of speech imprints and speech recognition. This & Zhang scale is applicable to China National Standard (CNS) A4 specification (210X297 mm) — " ~ (Please read the precautions on the back before filling this page) • Installation ·
、1T 經濟部智慧財產局員工消費合作社印製 -15- 514824 A7, 1T Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs -15- 514824 A7
五、發明説明（13) (請先閲讀背面之注意事項再填寫本頁) 在現實施例中，信號譯密裝置i 9加裝於輸入裝置 1〇中’以對發送於輸入裝置1 〇及電腦系統2 〇間之資 料譯密/解碼。同樣，譯密裝置2 5加裝於電腦系統2〇 中’其理由相同。譯密裝置1 9及2 5可爲在軟體控制下 之任何已知型式之譯密電路或處理器。譯密裝置1 9可與 介面控制器1 2整合，從而使用介面控制器1 2對資料訊 包化及譯密。 在電腦系統2 0中，輸入處理器接收來自輸入裝置 1 0之語首及鍵盤資料’並移交語音信號至語音辨識裝置 2 2。記憶器2 4顯示連接至語音副系統裝置2 2 ，用作 其儲存電路。輸出裝置2 3可爲電腦系統2 〇之一顯示裝 置，用以顯示語音副系統裝置2 2之輸出。I / 〇控制器 2 6在本實施例中顯示如電腦系統2 〇及電腦系統2 〇外 之貝料源’ s者如網際網路連接，電話服務，及周邊裝置間 之閘口介面。 經濟部智慧財產局員工消費合作社印製 在此構造中’語音輸入1 7及語音輸出1 8應合倂爲 一接收單位，用於線上電話。在此方面，激發網際網路提 供者（V〇I p )之語音。而且，可由使用介面卡，諸如 地區網路（L A N )卡，達成I /〇控制器2 6之功能。 在本實施例中，發音辨識系統用以提供一確實之裝置 ’以傳送語音及鍵擊信號自輸入裝置1 〇至電腦系統2〇 。任何窺察介面控制器1 2及輸入處理器2 1間之介面之 企圖接受譯密裝置1 9，2 5之作用。 現參考圖5 ，方塊圖顯示本發明之一實施例，其中， 本纸張尺度適用中.國國家標準（CNS ) A4規格（210X297公釐） -16- 514824 A7 _____B7_ 五、發明説明（14) _入裝置爲一語音辨識鍵盤1 〇。語音辨識鍵盤1 〇之機 箱槪要表示於1 1 ，內部具有介面控制器1 2 ，此接收來 (請先閱讀背面之注意事項再填寫本頁) 自鍵擊輸入1 3及語音處理器1 4之資料。語音處理器 1 4接收來自語音輸入1 7之語音信號，並連接至記憶器 16。 語音處理器1 4接收自語音輸入1 7進來之言語，作 爲語音信號，變換口述話句爲命令，適於經由介面控制器 1 2傳輸至貓腦系統2 0或電子裝置3 0。介面控制器 1 2合倂來自鍵擊輸入1 3及語音輸入1 4之資料，並選 擇發送其至電腦系統2 0及電子裝置3〇。 依據此實施例，語音辨識鍵盤1 0可用作普通鍵盤， 具有諸如U S B之一介面，以提供鍵擊輸入至電腦系統 2 0。而且，然而，來自鍵擊輸入之輸入可爲控制資料， 且可引導至電子裝置，諸如D V D或C D播放機，立體音 響，V C R，設定匣，或電視機，介面控制器1 2裝有介 面，諸如射頻（R F )或紅線（I R )，或U S B電纜， 如可與受控制之裝置3 0相容者。 經濟/部智慧財產局員工消費合作社印製 而且，處理經由語音輸入1 7所接收之口述命令爲其 鍵擊相等者，並發送其至介面控制器1 2。介面控制器處 理自語音處理器1 4所轉送之輸入，正如來自鍵擊輸入 1 3之資料，.發送其經一電腦介面，諸如一 U S B介面或 一 R F或I R鏈路而至電腦系統2 0，或經所需之任何其 他適當介面而至電子裝置3 0。故此，口述命令可遙控任 何型式之多媒體電子裝置，諸如D V D或C D，播放機， 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐Ί 514824 經濟部智慧財產局員工消費合作社印製 A7 ___B7_ 五、發明説明（15) 立體音響，V C R，設定匣，或電視機。 該組可能之鍵盤命令及輸入爲所有可能語音命令及輸 入之一副組。鍵盤恆具有既定之限制數目之鍵及鍵命令， 而語音命令及輸入則無實質上之限制。可有千計之語音命 令，而鍵盤輸入則通常限於數百命令。 本發明之一主要特色爲語音訊包在U S B電纜上發送 ，與變換語音命令爲鍵盤碼，及然後發送鍵盤碼於電纜上 不同。語音命令組不限於如過去之鍵盤碼。所支持之語音 命令之數僅受系統記憶器及先已記錄之命令數限制。語音 命令數可生長而無限制，因爲先記錄之語音命令可經由網 際網路下載新碼加以更新。當使用者進出由特定網站支持 之網站語音命令時，可目動下載不同之一組語音命令。 圖6顯示本發明之另一實施例，目的在提供直接進出 網際網路。在此實施例中，輸入裝置1 0在槪要表示之一 機箱1 1內具有介面控制器1 2連接至語音處理器1 4， 此連接至之記憶器1 6及語音輸入1 7。在此實施例中， 口述命令由語音處理器1 4變換爲對應資料，並發送其經 一介面至電腦系統2 0。電腦系統2 0構造連接至網際網 路。如此，可由語音發動進出網際網路。輸入裝置1 0可 爲一鍵盤，由語音發動網際網路進出，更換或補充鍵盤上 之熱鍵。此實.施例可變換語音命令爲電腦系統2 0所辨識 之命令，俾用於引導網際網路。可由本發明之此實施例進 出一世界性之電腦網路。 本發明另構想一方法，使用單個介面於語音及鍵擊資 本纸張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） (請先閱讀背面之注意事項再填寫本頁) .裝· 訂 -18- 514824 A7 ___B7 _ 五、發明説明（16) 料上。一電腦分離合倂之信號爲數位語音信號部份及鍵擊 信號部份，俾供分別處理。可執行語音印跡比較。本發明 包含由語音命令發動網際網路進出。此等語音命令可輸入 .至一鍵盤。並構想一方法，用以自一鍵盤由口述命令遙控 一電子裝置。 雖已參考較佳實施例，說明本發明，但精於本藝之人 士明瞭在本發明之精神及範圍內可有更改及修改。較佳實 施例之附圖及說明爲範例，而非限制本發明之範圍，且所 有此等更改及修改應涵蓋於本發明之精神及範圍內。 (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 本紙張尺度適用中國國家標準（CNS ) A4規格（210X297公釐） -19-V. Description of the invention (13) (Please read the notes on the back before filling in this page) In the present embodiment, the signal decryption device i 9 is added to the input device 1 ′ to send to the input device 1 0 and Data decryption / decoding in computer system. Similarly, the translation device 25 is installed in the computer system 20 'for the same reason. The cryptographic devices 19 and 25 may be any known type of cryptographic circuit or processor under software control. The decryption device 19 can be integrated with the interface controller 12 so as to use the interface controller 12 to encapsulate and decrypt data. In the computer system 20, the input processor receives speech prefixes and keyboard data 'from the input device 10 and transfers voice signals to the speech recognition device 22. The memory 2 4 is shown connected to the voice sub-system device 2 2 as its storage circuit. The output device 23 may be a display device of the computer system 20 for displaying the output of the voice sub-system device 22. In this embodiment, the I / O controller 26 displays, for example, computer systems 200 and computer systems 200, such as Internet connections, telephone services, and gateway interfaces between peripheral devices. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs In this structure, ‘voice input 17 and voice output 18 shall be combined into a receiving unit for online phone calls. In this regard, the voice of the Internet provider (VOIP) is stimulated. Furthermore, the function of the I / O controller 26 can be achieved by using an interface card, such as a local area network (LAN) card. In this embodiment, the pronunciation recognition system is used to provide a reliable device ′ for transmitting voice and keystroke signals from the input device 10 to the computer system 20. Any attempt to observe the interface between the interface controller 12 and the input processor 21 will accept the functions of the translation device 19, 25. Referring now to FIG. 5, a block diagram shows an embodiment of the present invention, in which the paper size is applicable. National National Standard (CNS) A4 specification (210X297 mm) -16- 514824 A7 _____B7_ V. Description of the invention (14) The access device is a speech recognition keyboard 10. The chassis of the voice recognition keyboard 1 〇 should be shown on 1 1, which has an interface controller 1 2 inside. This is received (please read the precautions on the back before filling this page). Key input 1 3 and voice processor 1 4 Information. The speech processor 14 receives a speech signal from the speech input 17 and is connected to the memory 16. The speech processor 14 receives the speech coming from the speech input 17 as a speech signal, and converts the spoken sentence into a command, which is suitable for transmitting to the cat brain system 20 or the electronic device 30 through the interface controller 12. The interface controller 12 combines the data from the key input 13 and the voice input 14 and selects to send it to the computer system 20 and the electronic device 30. According to this embodiment, the speech recognition keyboard 10 can be used as an ordinary keyboard with an interface such as USB to provide keystroke input to the computer system 20. Moreover, however, the input from the keystroke input can be control data and can be directed to an electronic device such as a DVD or CD player, stereo, VCR, setting box, or television, and the interface controller 12 is equipped with an interface, Such as radio frequency (RF) or red wire (IR), or USB cable, such as those compatible with the controlled device 30. Printed by the Consumers' Cooperative of the Ministry of Economy / Ministry's Intellectual Property Bureau. Moreover, the dictation command received via voice input 17 is processed as the keystroke equivalent, and it is sent to the interface controller 12. The interface controller processes the input forwarded from the voice processor 14 just like the data from the keystroke input 13 and sends it to the computer system via a computer interface, such as a USB interface or an RF or IR link. , Or via any other suitable interface required to the electronic device 30. Therefore, the spoken order can remotely control any type of multimedia electronic device, such as DVD or CD, player. ___B7_ V. Description of the invention (15) Stereo speakers, VCR, setting box, or television. This set of possible keyboard commands and inputs is one of all possible voice commands and inputs. The keyboard always has a predetermined limited number of keys and There are no substantial restrictions on voice commands and voice commands. There are thousands of voice commands, while keyboard input is usually limited to hundreds of commands. One of the main features of the present invention is that voice packets are sent on a USB cable, and Convert voice commands to keyboard codes, and then send the keyboard codes on the cable differently. The voice command set is not limited to the keyboard codes as in the past. The number of supported voice commands is limited only by the system memory and the number of previously recorded commands. Voice The number of commands can grow without limit, because the first recorded voice command can be updated by downloading a new code via the Internet When a user enters or exits a website voice command supported by a specific website, a different set of voice commands can be downloaded visually. FIG. 6 shows another embodiment of the present invention for providing direct access to the Internet. In this embodiment In the input device 10, there is an interface controller 12 in one of the cases 11 to be connected to the voice processor 14, which is connected to the memory 16 and the voice input 17. In this embodiment, The spoken command is converted into corresponding data by the speech processor 14 and sent to the computer system 20 via an interface. The computer system 20 is configured to connect to the Internet. In this way, voice can be launched into and out of the Internet. Input device 1 0 It can be a keyboard. Voice can be used to enter and exit the Internet. Replace or supplement the hot keys on the keyboard. This embodiment can change the voice command to a command recognized by the computer system 20 and can be used to guide the Internet. This embodiment of the present invention enters and exits a worldwide computer network. The present invention also contemplates a method that uses a single interface for voice and keystroke capital paper standards that apply the Chinese National Standard (CNS) A4 specification (2 10X297 mm) (Please read the precautions on the back before filling in this page). Binding · Order-18- 514824 A7 ___B7 _ V. Description of the invention (16) On the material. A computer separates the signal from the digital voice signal section The part and the keystroke signal part are processed separately. The voice imprint comparison can be performed. The present invention includes initiating Internet access by voice commands. These voice commands can be inputted to a keyboard. A keyboard remotely controls an electronic device by a spoken command. Although the present invention has been described with reference to the preferred embodiment, those skilled in the art understand that changes and modifications can be made within the spirit and scope of the invention. The drawings and descriptions are examples, rather than limiting the scope of the invention, and all such changes and modifications should be covered within the spirit and scope of the invention. (Please read the precautions on the reverse side before filling out this page) -Installed. 、-!! Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs This paper applies the Chinese National Standard (CNS) A4 specification (210X297 mm) -19-