201013635 九、發明說明： 【發明所屬之技術領域】 本發明係一種語音系統及方法，特別是涉及到一種智筹纽立 系統及方法。 〜'"曰 【先前技術】 目刖一些機器δ又備中设置之語音命令軟體都必須通過用戶點 擊熱鍵啟動軟體，然後說出命令，並且命令種類非常少，只能‘用 戶主動的輸入，沒有人機對話的感覺。 八b 比如，在使用的掌上型GPS裝置，大多還停留在手 命令階段^數亦支援語音輸人之裝置，也只能是用戶主動的輪 〇 入’並不能得到機器的相應語音回復。 【發明内容】 鑒於上述習知技術之缺點，本發明之主要 通過人機對話使機器設備執行語音輸入命 法，解決了人機對話不順暢的問題。卜之3 U統及方 為達成上述目的，本發明之技術方案為：一種智 實現今與機ϋ間更人性化與智慧化的對話，該方法的步^為：/ (1) 監聽與接收外部語音信號； μ (2) 辨識所接收到的語音信號； ❹ （3)根據賴識_語音錢，執行相應功能。 否與所立中’判斷所接收到的言吾音信號是 一語音資訊，不相符合則回饋回一語音信號轨丁下步驟雜不 步驟在Ϊ:二立，$一搜尋與語音信號相關的可執行功能 少鄉興邊音域相對應的可執行功能。 任立户⑵巾’亦包括一語音信號轉換步驟，將所接― 居音信號轉換成機器可辨識信號。 將所接收到的 祕通過多種語音及相關功能的設置，可讓使用者掩 的使用機器設備’對機器設備也更具親切感。使用者更便捷 5 201013635 本發明還提出了一種智彗注立 化與智慧化的觸，錢’實現人與機ϋ間更人性 二;S模组’可監聽與接收外部語音信號· ^ 辨識模組，其連接該監聽模組，二“ ’ 語音信號進行辨識； 、、、+該·模組所接收到的 一功能應用模組，其連接該 信號，執行相應功能。㈣辨識換組，可根據辨識到的語音 該辨識模組，還設置有一語 該辨識模組，另有—語音韓顧；1早70 ’存儲語音資料。 ο 機器可辨識的觸語音信號之_完成高級語音信號與 該辨識模組，設置有一判斷留：換。 是否在本系統中有存儲；還設置f音收到^音信號 語音信號、資訊提示。 9 口饋早兀，對外部給與 該功能應用模組，還設置有一搜 信號的對應功能並執行。 搜索早凡，搜哥已辨識的語音 用者術，本發明的智慧語音系統及方法，可方便使 【實; _更具人性化與智慧t ο 此技Ζίίΐίΐϊ具體實施例說明本發明之實施方式，熟， ΐΐ明書所揭示之内容輕易地瞭解本發明^ =童亦可f由其他不同之具體實例加以施行或 離本發明之精神下進行各種修飾與變更胤點與應用’在不悖 一種發明方法之_示意圖，如圖所示， 此以;“實現人與機器間更人性化與智慧化之對話， 步驟101 :監聽雜收外部語音健，在機器設備 動進入到監雜ϋ，監聽與接餅部語音錢，無需任何熱鍵觸 6 201013635 動’叫接㈣—語音啟崎令時，才對其 ====，_賴_識信號； 相符合，若相符合則為信H與所存儲的語音資料 “否，，進行步驟105 ; 步驟04 ; ^不相符合則為 步驟104 :根據所接收到的纽 破定為可勤m音信_ Hgfl： 目2魏；在辨識 Ο Ο 需機:執行相應操作，並在執^能 步驟105 :回饋回-語音信號。t使用者進订^回復； -種智統的模組結構示意圖，如圖所示， 該智慧語音祕使用上述方法，姆統包括：”日慧化的對話’ 一監聽模組1〇,可監聽與接收 逛可設置-語音啟動單元100，只有者該監聽模組10 動命令後，觸發該往音啟動田ί收到特別設置的一語音啟 的相應i作 動&⑽’機11柯執行其他語音命令 機利===:號首先’將接收到的高級χ語音信號轉換成 一判斷單元202,判定所接收到的語音作號β 存儲，如是則調用與該語音信號相對應的二:=:中有 如否則回饋-語音信號； 功4執行’ 一語音資料單元2G3 ’存儲語音資料，即為―⑦ 系統所要使用到的語音命令皆存儲於此單元中；本 置本 203中存儲的語音資料越多，即可使人機對話 系統的機器設備更具人性化；該語音資料單元203nm 201013635 ίίΐΐΐ進行設置載入’可重新置入新語音資料，也可刪除已 ^有-語音_單元 ’對外部 ，語音回饋單元 I對外單元，财對外語音信^==傳 出，並且所傳出的語音錢是通職語音轉鮮元， 識的底層信號轉換成高級語音信號。 、機了辨 叫能^用f:3二其連接該辨識模組20，可根據辨識到的 ^號^仃相應功能；該功能應用模組3〇，設置 疋302 ’使語音信號與相應功能相對應， > 葬早 ο 可執行實現相應功能；本關聯單元號即 可調整已有的關聯關係。 了叹置新的關聯關係，也 該功月b應用模組30，還設置有_ 4Φ - 〇Λ1 的語音信號的對應功能並執行，使機器更且 已辨識201013635 IX. Description of the Invention: [Technical Field] The present invention relates to a speech system and method, and more particularly to a smart system and method. ~'"曰[Prior Art] Seeing that some machine δ and the voice command software set in the middle must start the software by clicking the hotkey, and then speak the command, and the command type is very small, only the user's active input. There is no feeling of human-machine dialogue. For example, in the palm-type GPS devices used, most of them still stay in the hand command phase. The number also supports the device for voice input, and only the user's active wheel ’ does not get the corresponding voice reply of the machine. SUMMARY OF THE INVENTION In view of the above-mentioned shortcomings of the prior art, the present invention mainly solves the problem that the human-machine dialogue is not smooth by causing the machine device to perform a voice input method through human-machine dialogue. In order to achieve the above objectives, the technical solution of the present invention is: a smart realization of a more humanized and intelligent dialogue between the present and the machine, the steps of the method are: / (1) monitoring and receiving External voice signal; μ (2) Identify the received voice signal; ❹ (3) Perform the corresponding function according to the _ voice money. No, it is a voice message that is judged by the judgment. The voice signal is a voice message. If it does not match, it will be returned to a voice signal. The steps are not in the following steps: Erli, $1 search is related to the voice signal. Executable functions that correspond to the functions of the Shaoxiang Xingbian range. The Renlihu (2) towel also includes a voice signal conversion step to convert the connected voice signal into a machine identifiable signal. By setting the received secrets through a variety of voice and related functions, the user can hide the use of the machine equipment' to be more intimate with the machine equipment. The user is more convenient 5 201013635 The invention also proposes a touch of wisdom and wisdom, and the money 'realizes the humanity between the human and the machine; the S module can monitor and receive the external voice signal · ^ recognition mode The group is connected to the monitoring module, and the second "' voice signal is recognized; a function application module received by the module, which is connected to the signal, and performs corresponding functions. (4) Identification change group, According to the recognized speech, the identification module is further provided with a recognition module, and the other is - voice Han Gu; 1 early 70 'storage voice data. ο Machine identifiable touch voice signal _ complete advanced voice signal and Identification module, set a judgment to stay: change. Whether there is storage in the system; also set the f tone to receive the ^ tone signal voice signal, information prompt. 9 port feed early, the external application of the function application module, The corresponding function of the search signal is also set and executed. The search for the voice user, the intelligent voice system and method of the invention can be conveniently made [real; _ more humanized and intelligent t ο Technology 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 The change point and the application 'is not a schematic diagram of the invention method, as shown in the figure, this is; "to achieve a more humanized and intelligent dialogue between people and machines, step 101: monitor the miscellaneous external voice health, in The machine and equipment move into the monitoring, monitoring and voice of the cake, without any hot key touches, the call is made, the voice is called, and the voice is called. If there is a match, it is the letter H and the stored voice data "No, proceed to step 105; Step 04; ^ If it does not match, then step 104: According to the received Newton, it is a workable tone_ Hgfl: 目2魏; in the identification Ο 需 Need machine: perform the corresponding operation, and in the implementation step 105: feedback back-speech signal. t user subscribes to ^ reply; - a schematic diagram of the module structure, such as As shown in the figure, the smart voice secret uses the above The method includes: "Sun Huihua's dialogue", a monitoring module, a monitoring and receiving, and a voice-activated unit 100, and only the monitoring module 10 activates the command, triggering the voice to start the field. ί Receives a special setting of a voice-activated i-action & (10) 'machine 11 Ke performs other voice command arbitrage ===: number first 'converts the received advanced χ voice signal into a judgment unit 202, determines the received The obtained speech is stored as a number β, and if so, the second corresponding to the voice signal is called: =: there is an otherwise feedback-speech signal; the function 4 performs 'a voice data unit 2G3' to store the voice data, that is, the system requires The used voice commands are stored in the unit; the more voice data stored in the present 203, the more convenient the machine of the human-machine dialog system is; the voice data unit 203nm 201013635 ίίΐΐΐ is set to load 'You can re-input new voice data, you can also delete the existing voice-units to the external, voice feedback unit I external unit, the foreign voice message ^== outgoing, and the outgoing voice A through level speech-membered fresh, conversion identification signal into the underlying advanced voice signal. The machine can use the f:3 two to connect the identification module 20, according to the identified ^ number ^ 仃 corresponding function; the function application module 3 〇, set 疋 302 ' to make the voice signal and the corresponding function Correspondingly, > funeral early ο can implement the corresponding function; this associated unit number can adjust the existing relationship. The sigh is set to a new relationship, and the function module 30 is also applied to the power module b, and the corresponding function of the voice signal of _ 4Φ - 〇Λ1 is also set and executed to make the machine more recognized.
還有-語音載人模組4G ’其與辨識模组2G 2〇3相聯接，對該語音資料單元2〇3 的=2^ 載入新的語音資料。 仔儲的《"貝科進仃調整或 =地，將本發明系統置入一掌上型奶裝 $啟後，如需使用語音與GPS裝置對話， f 〇 咖’人齡綱Ϊ 用戶· Mantas。 發4ϊ動二：先並生 C 過回馈單元對用戶進行語音回饋。) 對接ίί的語進辨識模組20 模組^搜尋到語音信號相對應的可 較為堵車，另外-條走2環路，會經過南京路’但是 暢通，請問您要選擇哪-條呢？條多3公里，但是路徑 8 201013635 用戶：就選第二條吧。 絲_蝴好了，現咖直行，500Also, the voice manned module 4G' is coupled to the recognition module 2G 2〇3, and the new voice material is loaded to the voice data unit 2〇3. The storage of "" Beca 仃 仃 或 = = 将 仃 仃 仃 仃 仃 仃 本 本 本 本 本 本 本 本 本 本 本 本 本 本 本 仃 仃 仃 本 仃 本 本 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃 仃. Send 4 ϊ 2: First merge and C Pass the feedback unit to provide voice feedback to the user. ) The interface module 20 of the interface ί ^ 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻 搜寻More than 3 kilometers, but the path 8 201013635 User: Just choose the second one. Silk _ butterfly is good, now coffee straight, 500
Mantas ° 是的，王先生，我有什麼能為您效勞的嗎？ 我好困啊！ 用戶 系統 用戶 用戶 系統 系統：對林，姐聽清楚，能再說一次嗎？ 我真的好困啊！ 對不起，我沒聽清楚，能再說一次嗎？ Ο 用戶 關機 系統.0K ’關機進行中。 ㈣由知’本發明之智慧語音系統及方法，朗到機器設 機器更具人性化、智慧化，可讓使用者通過語音對 機器汉備進仃控制，與機器設備順暢對話。 铁。、f上所述’本發縣合發明專利之要件，爰依法提出專利申 二二二上所f者僅為本發明之較佳實施例’舉凡熟悉本發明 〇 嘗二以ΠΓ々由在爰依本發明精神所做之均等修飾或變化，皆應涵 盍於以下之申請專利範圍内。 【圖式簡單說明】 ί一圖係為本發明方法之流程示意圖。 第 rli係為本發明智慧語音系統之模組結構示意圖。 【主要元件符號說明】 監聽模塊 語音啟動單元 辨識模塊 語音轉換單元 判斷單元 語音資料單元 10 100 20 201 202 203 9 201013635 語音回饋單元 204 功能應用模組 30 搜索單元 301 關聯單元 302 語音輸入模組 40 101〜105 :流程步驟。 監聽與接收外部語音信號 101 將所接收到的語音信號轉換 102 成機器可辨識信號 判斷所接收到的語音信號是 103 〇 否與所存儲的語音資料相符合 根據所接收到的語音信號， 104 執行相應功能 回饋回一語音信號 105Mantas ° Yes, Mr. Wang, what can I do for you? I am so sleepy! User System User User System System: For Lin, sister can hear clearly, can you say it again? I am really sleepy! Sorry, I didn't hear it clearly. Can you say it again? Ο User Shutdown System .0K ’ Shutdown is in progress. (4) Knowing the intelligent voice system and method of the present invention, the machine is more user-friendly and intelligent, and allows the user to control the machine through the voice and smoothly communicate with the machine. iron. And the above-mentioned requirements of the 'Beifa County invention patent, 爰 提出 提出 专利 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 二 仅为 仅为 仅为 仅为 仅为 仅为 仅为 仅为 仅为 仅为 仅为 熟悉 熟悉 熟悉 熟悉 熟悉 熟悉Equivalent modifications or variations made in accordance with the spirit of the invention are intended to be within the scope of the following claims. [Simple description of the drawings] ί a diagram is a schematic flow chart of the method of the present invention. The rli is a schematic diagram of the module structure of the smart voice system of the present invention. [Main component symbol description] Monitoring module voice activation unit identification module voice conversion unit determination unit voice data unit 10 100 20 201 202 203 9 201013635 voice feedback unit 204 function application module 30 search unit 301 association unit 302 voice input module 40 101 ~105: Process steps. Listening and receiving the external voice signal 101 Converting the received voice signal 102 into a machine identifiable signal to determine whether the received voice signal is 103 〇 is in conformity with the stored voice data according to the received voice signal, 104 performing The corresponding function returns a voice signal 105