1303055 九、發明說明： 【發明所屬之技術領域】 一種產生對應資料的系統及其方法，特別是指一種利用互動 式語音回覆來產生對應資料的系統及其方法。 【先前技術】 隨著電子科技的發展，語音處理系統已在通訊裝置上提供各 式各樣的語音服務，所謂的語音服務指的是以電話號碼搭配互動 式語音回覆（Interactive Voice Response，IVR)系統來為撥入的 使用者進行各項服務的處理及查詢，例如航空公司的飛航資訊或 訂位、股票喊價系統、購物、帳務查詢等。 請參閱「第1圖」習知的語音服務系統之示意圖，當使用者 110以行動通訊裝置121撥打有提供互動式語音回覆系統“ο的 電話號碼時，行動通訊裝置121會透過電磁波與電話服務中心 130建立電信鏈結，並將使用者彳1〇的通訊引導到互動式語音回 復系、、先140，卩返之，使用者再藉由按壓行動通訊裝置HQ上之一 個按鍵或多個按鍵與互動式語音回覆系統14Q進行語音應答、識 別、接續、轉移等活動，以達成使用者110所需求之服務。所以， 互動式語音回I魏可達浙時服務使用者及料服務使用者的 ^事開鎖。、由於行動通訊裝置非ft及，故上軌行動通訊裝置 二胃'Ή吏用者110同樣可以使用一般的固定電話122來與 电話服務巾心13Q建立電錢結，進而得到其所需求之服務。 在电子產口口提供越來越多的娛樂功能下，人們在生活 1303055 中* Z越來越多樣化 歡唱歌，但往往^例如唱歌’不過，雖然人們喜 歌時因、 音域範圍的歌曲，因此在唱 /ΖΓ 低音超過自己能唱的範圍，以至於在高音時 定二理低音時也低不下來的情況。雖然大英百科全書有 二：士:音域範圍’如「第2A圖」之音域範圍表210所示， 有刀為男“、男Μ、男低音、女高音、 不過大英输#巾_谢謝齡終如ϋ =音域顧示意_所示，女高音與女中音只差兩個音高、 男南:料中音妓健只差了—個音高而已，所以對於大多數 人而吕’紋無法正確的騎出自己的音域制，產生無法選出 適合自己唱的歌關題。因此，如何能提供—種讓使用者可以方 便的得知自己的音域範圍所適合唱的歌曲的功能，成為大多數人 希望可以解決的問題之一。 【發明内容】 、鑒於以上的問題，本發明的主要目的在於提供—種利用互動 式語音回㈣統來產生音域㈣應資料的系統及其方法，透過互 動式語音回覆纽，讓使用者發峻音資料，進_斷使用者發 出語音資料的音域，並回應使用者對應其發出之音域的對應 料’讓使用者得知自己適合唱哪些歌曲，藉轉決先前技術所存 在之問題，達到讓使用者可以歡唱適合自己的歌曲的功效。 為達上述目的，本發明可以藉由方法與系統兩方面達成，本 發明所揭露之系統，包括有：通訊模組、接收模组、儲存模組、 1303055 資料庫模組、判斷模組、回應模組。本發明所揭露之 打列步驟：使用者撥號至互動式語音回覆系統、互較=括 覆系統接收由使用者發出之語音資料並儲存、判斷語音^曰: 域，並由㈣料讀出與錢者發出之語音資料之對、曰 應貧料、互動式語音回覆系統回應使用者對應資料Γ^：'’的對 有關本發日狀詳細舰與實作，驗合圖示在 細說明如下，其内容足以使任_f相難藝者了 _ = 術内容並據以實施，且根據本綱書所賊之崎及狀月之技 熟習相關技藝者可歸地轉本發_ M之 壬何 【實施方式】 ,2、下先卩第3圖」本發騎提之侧絲式語音 =應貧料之系統架構圖來說明本發明的系統運作。如圖所示 發明之系齡有通訊额310、接_@ 32q 判斷模Μ 340、資料細且· Γ 330、 31〇倉主奢、侧吴組350、回應模組，其中通訊模組 勒負二建立與制者之間㈣訊，使得制者發細聲音盘互 回覆系統所產生的語音可傳至對方；接收模組32〇負、 請她㈣撕細靖，細收到的 槪组33〇;儲存模，组330負責儲存接收模組咖 此η的。^曰貝料’判斷模組340負責由儲存模組330讀取接 語料料，觸接㈣語音㈣的音域，並由 35G處讀取對應於接收的語音龍的音域的對應資 貝庫极組350負責儲存對應於接收模組320所接收到的語 1303055 資=之曰域之對應資料；回應模組360負責將判斷模組由 者科庫模組·讀出的對應資料透過通訊· 31〇傳遞給使用 ^卜树_包含提示制者糾料:#解特輸 馬的提示模組370。1303055 IX. Description of the invention: [Technical field to which the invention pertains] A system for generating corresponding data and a method thereof, and more particularly to a system and method for generating corresponding data using an interactive voice response. [Prior Art] With the development of electronic technology, the voice processing system has provided a variety of voice services on the communication device. The so-called voice service refers to the telephone number and interactive voice response (IVR). The system is used to process and query various services for dial-in users, such as airline flight information or reservations, stock bidding system, shopping, and account inquiry. Please refer to the schematic diagram of the conventional voice service system in "FIG. 1". When the user 110 dials the telephone number provided by the mobile communication device 121 to provide the interactive voice response system, the mobile communication device 121 transmits the electromagnetic wave and the telephone service. The center 130 establishes a telecommunication link, and directs the user's communication to the interactive voice response system, first 140, and then the user presses a button or buttons on the mobile communication device HQ. The interactive voice response system 14Q performs voice response, identification, connection, transfer and other activities to achieve the service demanded by the user 110. Therefore, the interactive voice back I can reach the service of the Zhejiang time service user and the material service user. Since the mobile communication device is not ft, the upper-track mobile communication device 2 stomach user 110 can also use the general fixed telephone 122 to establish a money money knot with the telephone service towel core 13Q, thereby obtaining the Demand service. In the electronic production mouth to provide more and more entertainment features, people in life 1303055 * Z more and more diverse singing, but often ^ Such as singing 'However, although people like songs, the range of songs, so the sing / 低音 bass is more than the range that you can sing, so that when the treble is fixed, the bass is also low. Although the Encyclopedia There are two books in the book: Shi: The range of the range is as shown in the range of the range of 210 in the "A2A". There are knives for men, males, basses, sopranos, but the British loses #巾_Thank you for the end of your life. = The sound field Gu shows _, the soprano and the mezzo-soprano are only two pitches high, male south: the material mid-tone is only worse than a pitch, so for most people, Lu's pattern cannot be correct. Riding out your own range system, you can't choose the songs that you can sing. So how can you provide a function that allows users to easily know the range of songs that their range is suitable for singing? One of the problems that can be solved. SUMMARY OF THE INVENTION In view of the above problems, the main object of the present invention is to provide a system and method for generating a sound domain (4) application data by using an interactive voice back (four) system, through an interactive language Respond to the button, let the user send the sound data, enter the domain of the voice data sent by the user, and respond to the corresponding material of the user's corresponding sound field 'to let the user know which songs he is suitable to sing, by transferring the previous The problem of the technology is to achieve the effect that the user can sing the songs that are suitable for the user. In order to achieve the above object, the present invention can be achieved by the method and the system. The system disclosed by the present invention includes: a communication module The receiving module, the storage module, the 1303055 database module, the judging module, and the response module. The steps of the present disclosure are as follows: the user dials into the interactive voice response system, and the mutual comparison = coverage system receives The voice data sent by the user is stored and judged, and the voice is detected by the (4) material, and the voice data sent by the money is read, the poor respondent, and the interactive voice reply system respond to the user corresponding data. ''The detailed description of the ship and the actual ship, the check-in diagram is as follows, the content is enough to make the _f phase difficult to _ = the content of the operation and according to the implementation, and according to The skill of the thief of the thief and the skill of the genius in this program can be transferred to the local _ M 壬 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 【 2、 2、 2、 2、 2、 = System architecture diagram of the poor material to illustrate the operation of the system of the present invention. As shown in the figure, the age of the invention has a communication amount of 310, _@32q judgment module 340, data details and Γ 330, 31 warehouse main luxury, side Wu group 350, response module, in which the communication module is negative Second, establish (4) between the makers and the makers, so that the voice generated by the makers' voice-repetition system can be transmitted to the other party; the receiver module 32 is disappointing, ask her (4) to tear the details, and receive the group 33 carefully. 〇; storage module, group 330 is responsible for storing the receiving module. The 曰 料 ' ' judging module 340 is responsible for reading the linguistic material from the storage module 330, touching the (four) voice (four) range, and reading the corresponding vocabulary corresponding to the received voice dragon from 35G. The group 350 is responsible for storing corresponding data corresponding to the domain of the language 1303055 received by the receiving module 320; the response module 360 is responsible for transmitting the corresponding data of the determining module from the library module to the communication. 〇 Passed to the use of the ^ tree _ containing the prompt maker to correct: #解特马的提示模块 370.
亚可以在聽到「,」聲後開始哼唱語音資料，例如自由哼唱「生 曰快樂歌」或是跟著先前聽到崎音㈣哼唱，於是使用者發出 的5吾音貧料會透過通訊模組31〇建立的通訊傳送至接收模組 320 ’接收模，组320會把接收到的語音資料錄製為數位資料的格 、，接著[個實施例與來解說解說本翻的運作紐盘方法， ^參照「第4圖」本翻所提之利肢動式語音回覆檢測音感 护方法流賴。當使財欲本發_取得自“合唱的歌曲 百先要㈣話絲發日狀絲式語音回1魏，與本發明之 ^訊模組310建立通訊（步驟41G)，接著提示模組37G便會播 二請您、在『。畢』聲後輕鬆哼唱出—段自己拿手的曲調，哼唱結 /按#』子鍵。」的提示語，提示使用者哼唱聲音至互動式 回復纽’又或者提稍組训可以其他的提示語「想要有 ^幫你導唱-小段請按彳、想要自由哼唱翁2」，讓使用者選擇 是否要跟唱或者自由哼唱，如果使用者按下「彳」鍵選擇導唱，則 提示模組370會播放-段人聲的聲音資料，例如「祝你生日快樂、！ 祝如、生日快樂！…」在播放聲音資料之後，提示模組37〇會繼續 播放「請您在『，』_輕财唱，哼唱結束後請按『#』字鍵」， 不順使用者輸入「1」鍵或是「2」鍵，使用者都將聽到「,」聲， 1303055After hearing the "," sound, Ya can start to sing the voice data, such as freely singing "Happy Songs of Life" or following the previous sings of the singer (4), so the user sends out the 5th poor material through the communication mode. The communication established by the group 31〇 is transmitted to the receiving module 320 'receive mode, and the group 320 records the received voice data as a grid of digital data, and then [the embodiment and the explanation of the operation method of the flipping, ^ Refer to "4th picture". This method of fluent voice response detection sound sensation is reliant. When making the financial desire _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ You will broadcast the second, please sing it out after the "." sound - the tunes of your own hands, sing the knot / press the #』 subkey." prompts the user to sing the sound to the interactive reply button 'Or another group training can be other prompts "I want to have a help to guide you - please press 小, want to freely sing Weng 2", let the user choose whether to sing or sing freely, if When the user presses the "彳" button to select the guide, the prompt module 370 will play the sound data of the vocal, such as "Happy birthday to you, happy birthday, happy birthday!..." after playing the sound material, the prompt mode Group 37 will continue to play "Please sing in "," _ light sing, press 『#』字键 after humming, do not follow the user to enter "1" or "2", the user will Hear the sound of "," 1303055
式:並儲存到儲存模組330中（步驟440)，接著判斷模組34〇 會讀出使用者發出的語音㈣，朗斷使財哼唱的「生日快樂 歌」的音域’並讀取相對應的對應資料（步驟470)，如使用者所 °予唱的「生日快樂歌」的音域中最高音為「E3」、最低音2 , 判斷模=且_會根據「第5A圖」、「第5已圖」所示之音域範圍表 ⑽、音域範圍示意圖52〇，以最高音的「E3」判斷出使用者屬 ^男"音’並由綱模組35G中所記錄之對射料表6〇〇 弟6圖」）讀出屬於男中高音的第一歌名、第二歌名、第三歌 名’亚=回應模組360經由通訊模組31〇回應第一歌名、第二 3用給使用者知道（步驟480)。如此，本發明即可以 者砂自己的適合唱歌曲為第—歌名、第二歌名、第三歌 名’解決使用者無法選到合翻歌絲唱的問題。 料的ίΐίΓ触輯使时其所適合唱陳曲為對應資 名上也可以回應給使用者第-歌手、第二歌手的歌手 的:曲，1即=:::象所得知的歌手名稱找出該歌手所唱 組* 丁.…，=、…者U %的歌曲’因此回應使用者的對應資 料並不以歌曲名稱為限 2上彰㈣針，錢财纽動搞音 (步驟410)之後，万叙守兮立 m ,7Λ 動式—回覆系統會先播放歡迎語，並由 扣不枳、、且370播放讓使用者選 上 聲立，田声I主 * 曰或颁別的提示語「請輪入你的 曰力耳味备1，女聲或童聲續括9 , 里耳明备2，回主選單請按9，处Φ士主 备〇」，提示使用者輸入音域類 …束明 一、 接收核組320在接收到使用者 1303055 斷使用:=(步驟420)後’即可提供給判斷模組340在判 稱（烟7Q)，祕崎峨類別選出比 料^述之才同了唱的對應資料，例如選擇男聲，則獲得的對應資 __ 難'請綱最低音「B2」 第五歌名或是演二她資料將會是 _的播放模組 卞從用言了曰的语音貧料（步驟440)之後， 370可以「要馬上收聽你剛剛錄的聲音嗎？確定請按、以 ==提:語來提示使用者是否需要本發明播放先前所亨二 使用者按下「彳」鍵表示願意聽先前自己哼唱的歌曲 乂，播放她380便會域存麵33Q巾讀出鱗的語音資料並 透過通訊模組310播放給使用者（步驟450)。 、在儲存完使用者哼唱的語音資料（步驟44〇)後，使用者々 為先前対唱_程巾並不是外雜好，又或者在播放模Γ 380播放語音資料（步驟)後，使用者並不滿意自己哼唱的 結果時，提示模組370可以「是否滿意您的聲音？滿意請按，、 不滿意請按2」的提示語提示使用者按下「2」鍵來表示希望再重 新巧唱（步驟480) ’如此使用者即可再哼唱—次「生日快樂歌」， 並由接收模組320儲存至儲存模組33Q，當使用者滿意自已哼唱 的語音資料「生日快樂歌」之後，在步驟物中，判斷模址= 會將使用者第-次哼唱的語音資料與第二次哼唱的語音資料視為 10 1303055 同们k本進仃判斷，例如第二次哼唱時的最高音為％ 低曰為D2」，因為第一次哼唱時的最高音「低古 口曰沾厂η 布—一人了 :、「」’因此判斷模組會將使用者的最高音判斷為第二次哼 ”’而第二次哼唱的最低音「〇2」並未低於第—次哼口; 、、最低θ B2」’ g|此判斷模組判斷的最低音依然是「b2」，如 此，當使財選擇音域範圍為男聲時，判斷模、组340會判斷使用 者屬於「男向音」並由對應資料表_中讀出第六歌名的對應資 料’於是回應模、组360便會回應第六歌名給使用者。 … 、而$ 了增加判斷的準確度，本發明還有負責_與使用者建 立通撕’ ％景所產生的雜訊的強度的侧模組，在使用者 與互動式語音回覆系統建立通訊（步驟41〇)之後，偵測模組咖 就會_通訊中的雜訊的強度（步驟)，並提供判斷模組⑽ f步驟㈣巾可㈣敵财所發出的語音資射_訊，使判 斷模組340不會目為雜賴干擾而糊，造成仙者的困擾。 /另外，本發明也可以提供沒有提示模組37〇的語音互動回覆 f統’也就在使用者建立通訊（步驟41〇)之後，直接播放〜畢」 荦請使用者直接開料唱，如此可以減少熟悉本發_使用者在」 與本發明互動時，聽取的提示語的時間。 —频本㈣續述之較佳#施_露如上，財並非用以限 疋本發明，任何熟習相健藝者，在不麟本㈣之精神和範圍 内’所為之更賴麟’均屬本㈣之專娜魏圍，因此本發 明之專利保護綱翻本_#_之申料概_界定者^ 1303055 準。 【圖式簡單說明】 第1圖係習知之語音服務系統示意圖。 第2A圖係習知之音域範圍表。 第2B圖係習知之音域範圍示意圖。 第3圖係本㈣所提之_互動式語音贿產生對應資料之 系統架構圖。 第4圖係本發明所提之利用互動式語音回覆產生對應資料之 方法流程圖。 第5A圖係本發明實施例所提之音域範圍表。 第5B圖係本發明實施例所提之音域範圍示意圖。 第6圖係本發明實施例所提之對應資料表。 【主要元件符號說明】 110 使用者 121 行動通訊裝置 122 電話 130 電話服務中心 140 互動式語音回覆系統 210 音域範圍表 220 音域範圍 3〇〇 對應資料產生系統 310 通訊模組 12 1303055 320 接收模組 330 儲存模組 340 判斷模組 350 資料庫模組 360 回應模組 370 提示模組 380 播放模組 390 偵測模組 510 音域範圍表 520 音域範圍 600 對應資料表 步驟410 建立通訊 步驟420 接收音域類別 步驟430 偵測雜訊強度 φ 步驟440接收語音資料並儲存 步驟450播放語音資料 步驟460是否重發語音資料 步驟470判斷語音資料之音域並讀取對應資料 步驟480回應對應資料 13And stored in the storage module 330 (step 440), then the module 34 读出 will read the voice (4) issued by the user, and break the sound field of the "Happy Birthday Song" that the singer sings and read the phase Corresponding corresponding data (step 470), if the highest score of the "Happy Birthday Song" sung by the user is "E3", the lowest sound 2, the judgment mode = and _ will be based on "5A map", In the range of the range (10) and the range of the range of the sound range shown in Fig. 5, the highest sound "E3" is used to determine that the user belongs to the male "sound" and is recorded by the target module 35G. Table 6: Figure 6) The first song name, the second song name, and the third song name of the male middle and high notes are read. The sub-response module 360 responds to the first song name via the communication module 31. The second 3 is known to the user (step 480). Thus, the present invention can solve the problem that the user can't choose to sing and sing in the song-name, the second song name, and the third song name. The ΐ ΐ Γ Γ Γ 使 使 使 使 使 使 使 使 使 使 使 使 使 使 使 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈 陈Out of the singer's group * Ding...., =, ... U% of the song 'Therefore responding to the user's corresponding information is not limited to the name of the song 2 (4), the money is moving (step 410) After that, Wan Shou Guardian m, 7 Λ dynamic-reply system will play the welcome message first, and the 370 play will let the user choose the voice, Tian Sheng I master * 曰 or the prompt of the award "Please turn your enthusiasm for your ear, 1 female or child continuation, 9 for the ear, and 2 for the main menu." First, the receiving core group 320 receives the user 1303055 and uses it: = (step 420), then the judgment module 340 can be provided to determine (smoke 7Q), and the Mizusaki category selects the material. The corresponding information of the singer, for example, the choice of male voice, the corresponding capital obtained __ difficult 'please the lowest sound "B2" fifth song name or actress her data will After the _'s play module 卞 用 用 ( ( ( ( ( ( ( ( ( ( ( 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 370 The present invention plays the song that the previous user has pressed the "彳" button to indicate that he is willing to listen to the song he sang before, and then plays the 380 to read the voice data of the scale in the field storage area and broadcast it to the communication module 310. User (step 450). After storing the voice material sung by the user (step 44〇), the user sings that the previous singer _ the singer is not good, or after playing the voice data (step) in the playing module 380, If the user is not satisfied with the result of the singing, the prompt module 370 can "satisfy your voice? Please press, please press 2" prompts the user to press the "2" button to indicate that they wish to Re-spoken (step 480) 'So the user can sing again - "Happy Birthday Song", and stored by the receiving module 320 to the storage module 33Q, when the user is satisfied with the voice material "Happy Birthday" After the song, in the step, the judgment template = the user's first singer's voice data and the second singer's voice data will be regarded as 10 1303055. The highest note when singing is % low is D2", because the highest sound of the first singer is "Low Gukou 曰 厂 η — cloth - one person:, ""' so the judgment module will be the highest user The sound is judged as the second time 哼"' and the second humming "〇2" is not lower than the first pass; ,, the lowest θ B2"' g| The lowest note judged by this judgment module is still "b2", so when the profitable range is male, it is judged. The module and group 340 will judge that the user belongs to the "male tones" and read the corresponding data of the sixth song name from the corresponding data table _, and then respond to the mode, the group 360 will respond to the sixth song name to the user. ... and $ to increase the accuracy of the judgment, the present invention also has a side module responsible for establishing the strength of the noise generated by the user to tear through the '% scene, and establishing communication between the user and the interactive voice reply system ( After step 41), the detection module will _ the intensity of the noise in the communication (step), and provide the judgment module (10) f step (four) towel can (four) the voice signal issued by the enemy money, so that the judgment The module 340 will not be confused by the interference, causing confusion for the fairy. In addition, the present invention can also provide a voice interactive reply without the prompt module 37〇, and then directly play the video after the user establishes the communication (step 41〇), and asks the user to directly sing, It is possible to reduce the time to be familiar with the prompts that the user listens to when interacting with the present invention. - 频本(四)Continuously preferred #施_露上,财财 is not intended to limit the invention, any familiar artisan, in the spirit and scope of the 麟本(4) This (four) specializes in Wei Wei, so the patent protection of this invention is turned over _#_ The application is defined as ^ 1303055. [Simple description of the diagram] Fig. 1 is a schematic diagram of a conventional voice service system. Figure 2A is a table of ranges of conventional sound ranges. Figure 2B is a schematic diagram of the range of the known sound range. Figure 3 is a system architecture diagram of the corresponding data of the interactive speech bribery mentioned in (4). Figure 4 is a flow chart of a method for generating corresponding data by using an interactive voice response according to the present invention. Fig. 5A is a range of range of sound ranges mentioned in the embodiment of the present invention. FIG. 5B is a schematic diagram of the range of the sound field proposed by the embodiment of the present invention. Figure 6 is a correspondence table of the embodiments of the present invention. [Description of main component symbols] 110 User 121 Mobile communication device 122 Telephone 130 Telephone service center 140 Interactive voice response system 210 Range range table 220 Range range 3〇〇 Correspondence data generation system 310 Communication module 12 1303055 320 Receiver module 330 Storage module 340 judgment module 350 database module 360 response module 370 prompt module 380 play module 390 detection module 510 range range table 520 range 600 the corresponding data table step 410 establish communication step 420 receive range category step 430 Detecting the noise intensity φ Step 440: Receiving the voice data and storing step 450: Playing the voice data. Step 460: Resending the voice data. Step 470: Determining the voice field of the voice data and reading the corresponding data. Step 480: Responding to the corresponding data 13