201124949 六、發明說明： 【發明所屬之技術領域】 [0001] 本發明涉及一種聲音偵測系統及方法，尤其涉及一種聲 音偵測通報系統及方法。 【先前彳支術·】 [0002] 聽障者或視障者走在馬路上時，如果遇到四周有車輛對 其鳴按喇°八，或是旁人對其呼叫，他們可能無法及時察 覺，因此會帶來一些不便利之處，在安全上也是一大隱 患。 [0003] 目前市面上有可以循音辨位的電子設備，如聽從機器人 、跟隨機器人等服務型機器人，利用這些服務型機器人 可以幫助聽障者或視障者確定聲音來源的方向。但這些 服務型機器人價格昂貴，成本太高，使用起來也不方便 ，不能滿足實際需求。 [0004] 另外，目前市面上也存在有為視障者設計的盲人導航設 備，但這種盲人導航設備只能用來給使用者指引方位， 無法告知使用者周圍環境異常狀況的發生。 【發明内容】 [0005] 鑒於以上内容，有必要提供一種聲音偵測通報系統及方 法，其可運用於移動裝置中，即時偵測周圍環境的聲音 ，並在出現異常狀況時，向使用者發出預警資訊。 [0006] 一種聲音偵測通報系統，運行於移動裝置中，該移動裝 置中儲存有語音模型資料庫，該系統包括：聲音接收模 組，用於接收移動裝置的麥克風偵測到的聲音訊號，並 099100133 表單編號A0101 第4頁/共22頁 0992000280-0 201124949 計算該聲音訊號的能量；聲音辨位模組，用於當該聲音 訊號的能量大於預設值時，根據該聲音訊號的能量判定 該聲音來源的方向；聲音辨識模組，用於根據語音模型 資料庫中儲存的語音辨識的特徵參數及不同類別的聲音 模型，確定該聲音訊號的類別；通知模組，用於根據該 聲曰的類別發出預警貝sfl ’告知使用者該聲音來源的方 向。 [0007] Ο [0008] ❹ [0009] 一種聲音偵測通報方法，應用於移動裝置中，該移動裝 置中儲存有語音模型資料庫，該方法包括如下步驟：接 收移動裝置的麥克風偵測到的聲音訊冑，並#算該聲音 訊號的能量；當該聲音訊號的能量大於預設值時，根據 該聲音訊號的能量判定该聲音來源的方向；根據語音模 型資料庫中儲存的語音辨識的特徵參數及不同類別的聲 音模型，確定該聲音訊號的類別；根據該聲音的類別發 出預警資訊，告知使用者該聲音來源的方向。 相較於習知技術，所述的聲音偵測通報系統及方法，運 用於移動裝置中，吁以即時偵測周圍環境的聲音，並在 出現異常狀況時，向使用者發出預警資訊，從而以低成 本的方式實現了為聽障者或視障者提供安全預警的功能 ，提高了移動裝置的產品附加價值。 【實施方式】 參閱圖1所示，係本發明聲音偵測通報系統較佳實施方式 的應用環境圖。該聲a侦測通報系統12運行於移動裝置1 中，所述移動裝置1還包括語音模型資料庫11、中央處理 器13、電子羅盤14、麥克風15和顯示屏16。在本實施方 099100133 表單編號A0101 第5頁/共22頁 0992000280-0 201124949 [0010] ^中’所賴音_資料和料料語 特徵 =不同類:的聲音模型，其中，不同類別的聲音模 括，人聲杈型和車輛喇叭聲模型。 所述中央處理叫詩調騎述聲音_通報系統η， 並控制所料音_通報系㈣的執行。所述聲音_ 通報系統12⑽當料裝置1㈣後，取麥克風15 偵測到的周圍環境聲音，並在出現異常狀況時，向使用 者發出預警資訊。 [0011] 在本實施方式t所述電子羅盤14用於判斷聲音來源的 方向與移動裝置1之間的相對位置是否-致。如果兩者相 對位置—致’則判定移動裝置1的方向與使用者的方向-致’在判定聲音來源的方向時就以麥克風15為依據；如 果兩者相對位置不一致，目丨划中 个蚁則判疋移動裝置1的方向與使用 者的方向不-致，在判定聲音來源的方向時就以電子羅 盤14為依據，以減少錯誤判定㈣來源方向的機率。 [0012] 參閱圖2所示，係圖1中所示聲音翁測通報系統12的功能 模組圖。其巾，該聲音_通:㈣統12包括聲音接收模 組121、聲音辨位模組122、聲音辨識模組123和通知模 組124。本發明所稱的模組是完成—特定功能的電腦程式 段，比程式更適合於描述軟體在電腦中的執行過程，因 此在本發明以下對軟體描述都以模組描述。 所述聲音接收模組121用於接收麥克風15偵測到的聲音訊 號，並計算該聲音訊號的能量。在其他實施方式中，如 果麥克風15同時偵測到兩個（及以上）來自不同方位的 099100133 表單編號WHOl 0992000280-0 [0013] 201124949 不同聲音訊號，或同時偵測到兩個（及以上）來自同一 • 方位的不同聲音訊號時，聲音接收模組121將分別計算每 個聲音訊號的能量。 [0014] 所述聲音辨位模組1 2 2用於判斷該聲音訊號的能量是否大 於預設值，以及當該聲音訊號的能量大於預設值時，根 據該聲音訊號的能量判定該聲音來源的方向。在本實施 方式中，所述聲音訊號的能量是指該聲音訊號的振幅大 小0 0 [0015] 其中，聲音辨位模組122根據該聲音訊號的能量判定該聲 音來源的方向包括：獲取麥克風15的陣列中各點的聲音 訊號能量，然後比較麥克風15的陣列中各點的聲音訊號 能量大小，並根據預先設定的規則表判定該聲音來源的 方向。舉例而言，假設麥克風1 5的陣列為正三角形，參 閱圖3所示，麥克風15的陣列包括三個點：MIC1、MIC2 和MIC3。預先設定的規則表參閱圖4所示，如果點MIC1 的聲音訊號能量大於點MIC2的聲音訊號能量，且點MIC1 〇 的聲音訊號能量大於點MIC3的聲音訊號能量，則聲音辨 位模組122判定聲音來源的方向為前方。同理，根據圖4 中的規則表，可以判定聲音來源的其他方向，在此不再 贅述。 [0016] 在其他實施方式中，所述聲音辨位模組122根據該聲音訊 號的能量判定該聲音來源的方向還包括：當電子羅盤14 判定該聲音來源的方向與移動裝置1之間的相對位置不一 致時（即移動裝置1的方向與使用者的方向不一致），獲 取電子羅盤14的陣列中各點的聲音訊號能量，然後比較 099100133 表單編號A0101 第7頁/共22頁 0992000280-0 201124949 電子羅盤14的陣列中各點的聲音訊號能量大小，並根據 預先設定的規則表判定該聲音來源的方向。其中，利用 電子羅盤14的陣列判定該聲音來源的方向與利用麥克風 1 5的陣列判定該聲音來源的方向，方法是一樣的（參閱 圖3和圖4的描述），在此不再贅述。 [0017] 所述聲音辨識模組123用於根據語音模型資料庫11中儲存 的語音辨識的特徵參數及不同類別的聲音模型，確定該 聲音訊號的類別。具體而言，聲音辨識模組123先從語音 模型資料庫11獲取語音辨識的特徵參數，根據該特徵參 數對麥克風15偵測到的聲音訊號進行參數化。然後，聲 音辨識模組123將參數化後的聲音訊號與語音模型資料庫 11中儲存的聲音模型進行比對，以確定麥克風15偵測到 的聲音訊號的類別。其中，聲音訊號的類別包括：人聲 和車輛制Ά聲。 [0018] 在其他實施方式中，如果麥克風15同時偵測到兩個（及 以上）來自不同方位的不同聲音訊號，或同時偵測到兩 個（及以上）來自同一方位的不同聲音訊號時，聲音辨 識模組123將辨識出每個聲音訊號的類別。進一步地，如 果是人聲，可以利用現有的語音辨識技術分析出說話的 内容。 [0019] 所述通知模組124用於根據該聲音的類別發出預警資訊， 告知使用者該聲音來源的方向。具體而言，針對聽障者 ，通知模組124將首先透過移動裝置1的震動提醒使用者 注意，接著在顯示屏16上顯示該聲音的類別（如顯示人 形圖案或車輛的圖案）、說話的内容及該聲音來源的方 099100133 表單編號A0101 第8頁/共22頁 0992000280-0 201124949 [0020] ❹ 在本實施方式中，通知模組124可以根據移動裝置】是否 有啟動藍牙魏或是㈣⑽耳機減，來麟使用者 是聽障者還是視障者。如果㈣裝置i有啟動 藍牙功能或 〇 [0021] [0022] 向等針對視障者’通知模組124將首先透過移動裝置1 的震動提醒使用者注意，接著將聲音的類別（人聲或車 柄制八聲）、說話的内容及該聲音來源的方向等内容打 包成語音資訊’並將其傳送到使用者佩戴的耳機（藍牙 耳機或有線耳機）上，透過耳機將該語音資訊播放出來 (如後方有車輛鳴喇„八）。 ”有線耳機相連H知模組124判定使用者視障者；如 果移動裝置1沒有啟動藍牙功能，且沒有與有線耳機相連 則通去模組124判錢用者聽障者^在其他實施方式中 也可以預先在移動裳置丨中設定使用模式（聽障者模式 或視障者模式），之後，當該聲音訊號的能量大於預設 值時，通知模組124將獲取該預先設定的使用模式 ，並根 據預先設定的使用模式發出預警資訊。 在其他實施方式中，如果麥克風15同時㈣到兩個（及 以上）來自不同方位的不同聲音訊號，或同時偵測到兩 個（及以上）來自同—方位的不同聲音訊號時，通知模 _根據聲音辨識模組胸識出的每個 的 類別，發出不同的預警資訊。 參閱圖5所示 的流程圖。 係本發明聲音偵測通報核較佳實施方式 [0023] 步驟S卜聲音魏模組121接收麥克_貞測到的聲音訊 099100133 表單編號A0101 第9頁/共22頁 0992000280-0 201124949 號，並計算該聲音訊號的能量。在其他實施方式中，如 果麥克風15同時偵測到兩個（及以上）來自不同方位的 不同聲音訊號，或同時偵測到兩個（及以上）來自同一 方位的不同聲音訊號時，聲音接收模組121將分別計算每 個聲音訊號的能量。 [0024] 步驟S2，聲音辨位模組122判斷該聲音訊號的能量是否大 於預設值。如果該聲音訊號的能量大於預設值，執行步 驟S3 ;如果該聲音訊號的能量小於或等於該預設值，返 回步驟S1。 [0025] 步驟S3，聲音辨位模組122根據該聲音訊號的能量判定該 聲音來源的方向。在本實施方式中，所述聲音訊號的能 量是指該聲音訊號的振幅大小。 [0026] 其中，聲音辨位模組122根據該聲音訊號的能量判定該聲 音來源的方向包括：獲取麥克風15的陣列中各點的聲音 訊號能量，然後比較麥克風15的陣列中各點的聲音訊號 能量大小，並根據預先設定的規則表判定該聲音來源的 方向。具體過程參見圖3和圖4的描述。 [0027] 在其他實施方式中，所述聲音辨位模組122根據該聲音訊 號的能量判定該聲音來源的方向還包括：當電子羅盤14 判定該聲音來源的方向與移動裝置1之間的相對位置不一 致時（即移動裝置1的方向與使用者的方向不一致），獲 取電子羅盤14的陣列中各點的聲音訊號能量，然後比較 電子羅盤14的陣列中各點的聲音訊號能量大小，並根據 預先設定的規則表判定該聲音來源的方向。其中，利用 099100133 表單編號A0101 第10頁/共22頁 0992000280-0 201124949 [0028] Ο [0029]201124949 VI. Description of the Invention: [Technical Field] The present invention relates to a sound detection system and method, and more particularly to a sound detection notification system and method. [Previous sputum surgery] [0002] When a hearing impaired person or visually impaired person walks on the road, they may not be able to detect it in time if they encounter a vehicle that is slamming around them or calling others to call them. Therefore, it will bring some inconveniences, and it is also a big hidden danger in terms of safety. [0003] There are currently electronic devices on the market that can be used for voice recognition, such as service robots such as robots and followers. These service robots can help the hearing impaired or visually impaired to determine the direction of the sound source. However, these service robots are expensive, costly, and inconvenient to use, and cannot meet actual needs. [0004] In addition, there are currently blind navigation devices designed for the visually impaired, but such blind navigation devices can only be used to guide the user to the orientation, and cannot inform the user of the occurrence of abnormal environment. SUMMARY OF THE INVENTION [0005] In view of the above, it is necessary to provide a sound detection notification system and method, which can be used in a mobile device to instantly detect the sound of the surrounding environment and send the user to the user in the event of an abnormal situation. Early warning information. [0006] A sound detection notification system is run in a mobile device, wherein the mobile device stores a voice model database, and the system includes: a sound receiving module, configured to receive an audio signal detected by a microphone of the mobile device, And 099100133 Form No. A0101 Page 4 / Total 22 Pages 0992000280-0 201124949 Calculate the energy of the sound signal; the sound discerning module is configured to determine the energy of the sound signal according to the energy of the sound signal when the energy of the sound signal is greater than a preset value The direction of the sound source; the sound recognition module is configured to determine the type of the sound signal according to the feature parameters of the voice recognition stored in the voice model database and the sound models of different categories; and the notification module is configured to The category issued an alert to the shell sfl 'inform the user of the direction of the source of the sound. [0007] 0008 [0009] A method for sound detection notification is applied to a mobile device, wherein the mobile device stores a voice model database, and the method includes the following steps: receiving a microphone detected by the mobile device Sound signal, and # calculate the energy of the sound signal; when the energy of the sound signal is greater than a preset value, determine the direction of the sound source according to the energy of the sound signal; according to the characteristics of the voice recognition stored in the voice model database The parameter and the sound model of different categories determine the type of the sound signal; and issue an early warning message according to the category of the sound to inform the user of the direction of the sound source. Compared with the prior art, the sound detection notification system and method are applied to a mobile device to instantly detect the sound of the surrounding environment and provide an early warning information to the user when an abnormal situation occurs, thereby A low-cost way to provide a security alert for the hearing impaired or visually impaired, increasing the added value of mobile device products. [Embodiment] Referring to Figure 1, there is shown an application environment diagram of a preferred embodiment of the sound detection notification system of the present invention. The sound a detection notification system 12 operates in the mobile device 1, and the mobile device 1 further includes a voice model database 11, a central processor 13, an electronic compass 14, a microphone 15, and a display screen 16. In the present embodiment 099100133 Form No. A0101 Page 5 / Total 22 Pages 0992000280-0 201124949 [0010] ^ In the sound of 'sound_ data and material features = different categories: the sound model, where different categories of sound mode Including, vocal squats and vehicle horn sound models. The central processing is called the poetry riding sound_notification system η, and controls the execution of the sounding_notification system (4). The sound_notification system 12(10) takes the ambient sound detected by the microphone 15 after the device 1 (4), and sends an early warning message to the user when an abnormal situation occurs. [0011] The electronic compass 14 in the present embodiment t is used to determine whether or not the relative position between the direction of the sound source and the mobile device 1 is correct. If the relative position of the two is opposite, the direction of the mobile device 1 and the direction of the user are determined to be based on the microphone 15 when determining the direction of the sound source; if the relative positions of the two are inconsistent, the target is selected. Then, it is judged that the direction of the mobile device 1 is not in the direction of the user, and when determining the direction of the sound source, the electronic compass 14 is used as a basis to reduce the probability of the erroneous determination (4) source direction. [0012] Referring to FIG. 2, it is a functional block diagram of the sound transmission notification system 12 shown in FIG. 1. The towel, the voice: (4) system 12 includes a voice receiving module 121, a voice recognition module 122, a voice recognition module 123, and a notification module 124. The module referred to in the present invention is a computer program that performs a specific function, and is more suitable for describing the execution process of the software in the computer than the program. Therefore, the following description of the software in the present invention is described by a module. The sound receiving module 121 is configured to receive the sound signal detected by the microphone 15 and calculate the energy of the sound signal. In other embodiments, if the microphone 15 simultaneously detects two (and above) 099100133 form numbers from different directions, WHO1 0992000280-0 [0013] 201124949 different sound signals, or simultaneously detects two (and above) from When different sound signals of the same direction are used, the sound receiving module 121 will calculate the energy of each sound signal separately. [0014] The sound discerning module 1 2 2 is configured to determine whether the energy of the sound signal is greater than a preset value, and when the energy of the sound signal is greater than a preset value, determine the sound source according to the energy of the sound signal. The direction. In the present embodiment, the energy of the sound signal refers to the amplitude of the sound signal. The voice discrimination module 122 determines the direction of the sound source according to the energy of the sound signal, including: acquiring the microphone 15 The sound signal energy of each point in the array is then compared with the sound signal energy of each point in the array of the microphone 15, and the direction of the sound source is determined according to a preset rule table. For example, assuming that the array of microphones 15 is an equilateral triangle, as shown in Figure 3, the array of microphones 15 includes three points: MIC1, MIC2, and MIC3. As shown in FIG. 4, if the sound signal energy of the point MIC1 is greater than the sound signal energy of the point MIC2, and the sound signal energy of the point MIC1 〇 is greater than the sound signal energy of the point MIC3, the sound discrimination module 122 determines The direction of the sound source is forward. Similarly, according to the rule table in Fig. 4, other directions of the sound source can be determined, and will not be described here. [0016] In other embodiments, determining, by the sound recognition module 122, the direction of the sound source according to the energy of the sound signal further includes: when the electronic compass 14 determines the relative direction between the sound source and the mobile device 1 When the positions are inconsistent (that is, the direction of the mobile device 1 does not coincide with the direction of the user), the sound signal energy of each point in the array of the electronic compass 14 is acquired, and then the comparison is 099100133. Form No. A0101 Page 7 / Total 22 Pages 0992000280-0 201124949 The sound signal energy of each point in the array of compasses 14 is determined according to a preset rule table. Wherein, the direction of the source of the sound is determined by the array of the electronic compass 14 and the direction of the source of the sound is determined by the array of the microphones 15. The method is the same (refer to the description of FIG. 3 and FIG. 4), and details are not described herein again. [0017] The sound recognition module 123 is configured to determine the type of the sound signal according to the feature parameters of the voice recognition stored in the voice model database 11 and the sound models of different categories. Specifically, the voice recognition module 123 first acquires the feature parameters of the voice recognition from the voice model database 11, and parameterizes the voice signals detected by the microphone 15 according to the feature parameters. Then, the sound recognition module 123 compares the parameterized sound signal with the sound model stored in the voice model database 11 to determine the type of the sound signal detected by the microphone 15. Among them, the categories of audio signals include: vocals and vehicle hum. [0018] In other embodiments, if the microphone 15 detects two (or more) different sound signals from different directions at the same time, or simultaneously detects two (and above) different sound signals from the same direction, The voice recognition module 123 will recognize the category of each voice signal. Further, if it is a human voice, the existing speech recognition technology can be used to analyze the content of the speech. [0019] The notification module 124 is configured to send an early warning information according to the type of the sound, and inform the user of the direction of the sound source. Specifically, for the hearing impaired, the notification module 124 will first alert the user through the vibration of the mobile device 1, and then display the category of the sound on the display screen 16 (such as displaying a humanoid pattern or a pattern of the vehicle), speaking. Content and the source of the sound 099100133 Form No. A0101 Page 8 / Total 22 Pages 0992000280-0 201124949 [0020] 本 In the present embodiment, the notification module 124 can activate the Bluetooth Wei or (4) (10) headphones according to the mobile device. Reduce, whether the user is a hearing impaired or a visually impaired. If (4) device i has a Bluetooth function enabled or 〇 [0021] [0022] The notification module 124 for the visually impaired person will first alert the user through the vibration of the mobile device 1, and then the category of the voice (vocal or handle) The contents of the eight sounds, the content of the speech, and the direction of the source of the sound are packaged into voice information 'and transmitted to the earphone (Bluetooth earphone or wired earphone) worn by the user, and the voice information is played through the earphone (eg There is a vehicle 鸣 八 八 ”.” The wired headset is connected to the H-known module 124 to determine the visually impaired person; if the mobile device 1 does not activate the Bluetooth function and is not connected to the wired headset, the module 124 is passed to the user. In other embodiments, the usage mode (the hearing impaired mode or the visually impaired mode) may be set in advance in the mobile mobile device, and then, when the energy of the audio signal is greater than the preset value, the notification module is notified. 124 will acquire the preset usage mode and issue an early warning message according to a preset usage mode. In other embodiments, if the microphone 15 simultaneously (four) to two (and above) different sound signals from different directions, or simultaneously detect two (and above) different sound signals from the same direction, the notification mode _ Different warning information is issued according to each category of the voice recognition module. See the flow chart shown in Figure 5. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT OF THE INVENTION [0023] Step S Bu sound module 121 receives the microphone _ 贞 measured voice 099100133 Form No. A0101 Page 9 / Total 22 pages 0992000280-0 201124949, and Calculate the energy of the sound signal. In other embodiments, if the microphone 15 detects two (or more) different sound signals from different directions at the same time, or simultaneously detects two (and above) different sound signals from the same direction, the sound receiving mode Group 121 will calculate the energy of each sound signal separately. [0024] Step S2, the sound discrimination module 122 determines whether the energy of the sound signal is greater than a preset value. If the energy of the sound signal is greater than the preset value, step S3 is performed; if the energy of the sound signal is less than or equal to the preset value, the process returns to step S1. [0025] In step S3, the sound discrimination module 122 determines the direction of the sound source based on the energy of the sound signal. In this embodiment, the energy of the audio signal refers to the amplitude of the sound signal. The sound discriminating module 122 determines the direction of the sound source according to the energy of the sound signal, including: acquiring sound signal energy of each point in the array of the microphone 15, and then comparing the sound signals of each point in the array of the microphone 15 The amount of energy, and the direction of the sound source is determined according to a preset rule table. For the specific process, refer to the description of FIG. 3 and FIG. 4. [0027] In other embodiments, determining, by the sound recognition module 122, the direction of the sound source according to the energy of the sound signal further comprises: when the electronic compass 14 determines the relative direction between the sound source and the mobile device 1 When the positions are inconsistent (that is, the direction of the mobile device 1 is inconsistent with the direction of the user), the sound signal energy of each point in the array of the electronic compass 14 is acquired, and then the sound signal energy of each point in the array of the electronic compass 14 is compared, and according to A preset rule table determines the direction of the sound source. Among them, use 099100133 form number A0101 page 10 / total 22 pages 0992000280-0 201124949 [0028] Ο [0029]
GG
[0030] 099100133 電子羅盤14的陣關定該聲音來源的方向與利用麥克風 15的陣列判定該聲音來源的方向，方法是—樣的（參閲 圖3和圖4的描述），在此不再贅述。 步驟S4，聲音辨識模組123根據語音模型資料庫η中儲存 的語音辨識的特徵參數及不同類別的聲音模型，確定該 聲音訊號的類別。具體而言，聲音辨識模組123先從語音 模型資料庫11獲取語音辨識的特徵參數，根據該特徵參 數對麥克風15偵測到的聲音訊號進行參數化。然後，聲 音辨識模組123將參數化後的聲音訊號與語音模型資料庫 11中儲存的聲音模型進行比對，以破定麥克風1 5摘測到 的聲音訊號的類別。其中，聲音訊號的類别包括：人聲 和車輛喇η八聲。 在其他實施方式中，如果麥克風15同時偵測到兩個（及 以上）來自不同方位的不同聲音訊號，或同時偵測到兩 個（及以上）來自同一方位的不同聲音訊號時，聲音辨 識模組123將辨識出每個聲音訊號的類別。進一步地，如 果疋人聲’可以利用現有的語音辨識技術分析出說話的 内容。 步驟S5 ’通知模組！ 24根據該聲音的類別發出預警資訊， σ知使用者該聲音來源的方向。具體而言，針對聽障者 ’通知模組124將首先透過移動裝置1的震動提醒使用者 、、主音 'ν’接著在顯示屏16上顯示該聲音的類別（如顯示人 形圖案或車輛的圖案）、說話的内容及該聲音來源的方 向等。針對視障者，通知模組124將首先透過移動裝置i 的震動提醒使用者注意，接著將聲音的類別（人聲或車 表單編號A0101 第11頁/共22頁 0992000280-0 201124949 輛喇叭聲）、說話的内容及該聲音來源的方向等内容打 包成語音資訊，並將其傳送到使用者佩戴的耳機（藍牙 耳機或有線耳機）上，透過耳機將該語音資訊播放出來 (如後方有車輛鳴喇°八）。 [0031] 在本實施方式中，通知模組124可以根據移動裝置1是否 有啟動藍牙功能或是否與有線耳機相連，來判斷使用者 是聽障者還是視障者。如果移動裝置1有啟動藍牙功能或 與有線耳機相連，則通知模組124判定使用者視障者；如 果移動裝置1沒有啟動藍牙功能，且沒有與有線耳機相連 ，則通知模組124判定使用者聽障者。在其他實施方式中 ，也可以預先在移動裝置1中設定使用模式（聽障者模式 或視障者模式），之後，當該聲音訊號的能量大於預設 值時，通知模組124將獲取該預先設定的使用模式，並根 據預先設定的使用模式發出預警資訊。 [0032] 在其他實施方式中，如果麥克風15同時偵測到兩個（及 以上）來自不同方位的不同聲音訊號，或同時偵測到兩 個（及以上）來自同一方位的不同聲音訊號時，通知模 組124將根據聲音辨識模組123辨識出的每個聲音訊號的 類別，發出不同的預警資訊。 [0033] 在本實施方式中，所述移動裝置1可以為手機、數碼相機 或PDA (Personal Digital Assistant，個人數位助 理）等具有資料處理功能的電子設備。 [0034] 最後應說明的是，以上實施方式僅用以說明本發明的技 術方案而非限制，儘管參照較佳實施方式對本發明進行 099100133 表單編號A0101 第12頁/共22頁 0992000280-0 201124949 了詳細說明，本領域的普通技術人員應當理解，可以對 本發明的技術方案進行修改或等同替換，而不脫離本發 明技術方案的精神和範圍。 【圖式簡單說明】 [0035] 圖1係本發明聲音偵測通報系統較佳實施方式的應用環境 圖。 [0036] 圖2係圖1中所示聲音偵測通報系統12的功能模組圖。 [0037] 圖3係圖1中所示麥克風1 5的陣列示意圖。[0030] 099100133 The array of electronic compasses 14 determines the direction of the sound source and determines the direction of the source of the sound using the array of microphones 15, as is the case (see the description of Figures 3 and 4), no longer Narration. In step S4, the voice recognition module 123 determines the type of the voice signal based on the feature parameters of the voice recognition stored in the voice model database η and the sound models of different categories. Specifically, the voice recognition module 123 first acquires the feature parameters of the voice recognition from the voice model database 11, and parameterizes the voice signals detected by the microphone 15 according to the feature parameters. Then, the sound recognition module 123 compares the parameterized sound signal with the sound model stored in the voice model database 11 to break the type of the sound signal extracted by the microphone 15. Among them, the categories of sound signals include: vocals and vehicles. In other embodiments, if the microphone 15 detects two (or more) different sound signals from different directions at the same time, or simultaneously detects two (and above) different sound signals from the same direction, the sound recognition mode Group 123 will recognize the category of each audio signal. Further, if the human voice can use the existing speech recognition technology to analyze the content of the speech. Step S5 'Notify the module! 24 issuing an early warning message according to the type of the sound, and knowing the direction of the user's source of the sound. Specifically, the notification module 124 for the hearing impaired will first alert the user through the vibration of the mobile device 1, and the main tone 'ν' then displays the category of the sound on the display screen 16 (such as displaying a humanoid pattern or a pattern of the vehicle). ), the content of the speech and the direction of the source of the sound. For the visually impaired, the notification module 124 will first alert the user through the vibration of the mobile device i, and then the type of the sound (vocal or car form number A0101 page 11 / total 22 pages 0992000280-0 201124949 horn sound), The content of the speech and the direction of the source of the sound are packaged into voice information, and transmitted to the earphone (Bluetooth earphone or wired earphone) worn by the user, and the voice information is played through the earphone (if there is a vehicle in the rear) ° eight). [0031] In the embodiment, the notification module 124 can determine whether the user is a hearing impaired or a visually impaired person according to whether the mobile device 1 has a Bluetooth function enabled or is connected to a wired headset. If the mobile device 1 has a Bluetooth enabled function or is connected to a wired headset, the notification module 124 determines that the user is visually impaired; if the mobile device 1 does not activate the Bluetooth function and is not connected to the wired headset, the notification module 124 determines the user Hearing impaired. In other embodiments, the usage mode (the hearing impaired mode or the visually impaired mode) may be set in advance in the mobile device 1, and then, when the energy of the audio signal is greater than a preset value, the notification module 124 will acquire the Pre-set usage mode, and issue warning information according to the preset usage mode. [0032] In other embodiments, if the microphone 15 detects two (or more) different sound signals from different directions at the same time, or simultaneously detects two (and above) different sound signals from the same direction, The notification module 124 will issue different warning information according to the type of each audio signal recognized by the voice recognition module 123. [0033] In the present embodiment, the mobile device 1 may be an electronic device having a data processing function such as a mobile phone, a digital camera, or a PDA (Personal Digital Assistant). [0034] Finally, it should be noted that the above embodiments are only for explaining the technical solutions of the present invention and are not limited thereto, although the present invention is carried out with reference to the preferred embodiments of 099100133 Form No. A0101 Page 12 / Total 22 Pages 0992000280-0 201124949 The detailed description of the technical solutions of the present invention may be made without departing from the spirit and scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS [0035] FIG. 1 is an application environment diagram of a preferred embodiment of a sound detection notification system of the present invention. 2 is a functional block diagram of the sound detection notification system 12 shown in FIG. 1. 3 is a schematic diagram of an array of microphones 15 shown in FIG. 1.
[0038] 圖4係本發明中判定聲音來源的方向的規則表。 [0039] 圖5係本發明聲音偵測通報方法較佳實施方式的流程圖。 【主要元件符號說明】 [0040] 移動裝置：1 [0041] 語音模型資料庫：11 [0042] 聲音偵測通報系統：1 2 [0043] 中央處理器：13 [0044] 電子羅盤：14 [0045] 麥克風：15 [0046] 顯示屏：1Θ [0047] 聲音接收模組：121 [0048] 聲音辨位模組：122 [0049] 聲音辨識模組：123 099100133 表單編號A0101 第13頁/共22頁 0992000280-0 124 201124949 [0050] 通知模組 099100133 表單編號A0101 第14頁/共22頁 0992000280-04 is a rule table for determining the direction of the sound source in the present invention. 5 is a flow chart of a preferred embodiment of the sound detection notification method of the present invention. [Main component symbol description] [0040] Mobile device: 1 [0041] Speech model database: 11 [0042] Sound detection notification system: 1 2 [0043] Central processing unit: 13 [0044] Electronic compass: 14 [0045] Microphone: 15 [0046] Display: 1Θ [0047] Sound receiving module: 121 [0048] Sound recognition module: 122 [0049] Sound recognition module: 123 099100133 Form number A0101 Page 13 of 22 0992000280-0 124 201124949 [0050] Notification Module 099100133 Form No. A0101 Page 14 / Total 22 Pages 0992000280-0