200425766 1 12l8twf. doc/006 屬之技術領域 本發明是有關於一種耳機，且特別是有關於一種反饋 式主動噪音控制耳機用之反饋式主動噪音控制電路。 先i技術 在電氣化產品日益普及的今日，音響器材已是人們消 遣娛樂與獲取新知之重要設備，其中，耳機更是提供隨時 隨地收聽之便利器具。爲了提供較佳之收聽效果，必須對 於耳機使用者會同時收聽到之環境噪音，採取噪音防制方 法。而依據採用的噪音防制方法之不同，可將耳機槪分爲 兩種··被動噪音防制耳機及主動噪音控制耳機。 被動卩呆曰防制耳機由於只是單純地依賴隔聲材料來 降低環境噪音，所以抗噪音的能力便與所使用材料的材 質、厚度、結構設計、貼合性、等有極大的關連，以致此 種耳機一般而言均極大型且厚重。此外，因使用在被動噪 音防制耳機上的材料，對於低頻的噪音幾乎沒有阻隔的能 力，以致如引擎、鼓風機之類的低頻噪音，就幾乎沒有噪 音防制的效果。反之，主動噪音控制耳機就沒有上述的限 制，因此，乃十分受到消費者的喜愛。 然而，一般市售的主動噪音控制耳機，通常只在其左 右喇叭前方各放置了一個麥克風感測器。此種作法，無= 其放置位置爲何，均只能靠一個麥克風感測器來接收喇叭 前方之噪音訊號，以致對於麥克風感測器本身性能的要求 就非常高，除了要挑選靈敏度高且價格昂貴的麥克風=測 器之外’甚至在生產裝配時，更爲了確保麥克風感測^仍 200425766 1 1 2 1 8tvvf.doc/006 可保持原有的高靈敏度，使得量產時，焊接過程甚易傷及 麥克風感測器，而影響生產的良率及成本。再者，一般反 饋式主動噪音控制耳機的麥克風感測器均放置於喇叭前方 0^〜lcm的距離內，會產生嚴重的近場效應，故即使使用高 靈敏度之麥克風感測器，仍然會受到喇叭前方近場效應的 影響，導致降噪效果大打折扣。 此外，習知之反饋式主動噪音控制耳機中的主動噪音 控制電路，因未能考慮將放音機等音樂發聲裝置所產生之 音訊輸入訊號，與麥克風感測器感測環境噪音所得之噪音 感測δΛ 5虎的增益調整電路分離，導致爲了抗噪效果而調整 噪音感測訊號之增益時，會連帶影響音樂原來頻譜，進而 可能產生低頻破音，或當使用者聽音樂而打開主動噪音控 制電路之電源時，音樂的音量會突然變大，造成耳朵極^ 舒服之情形。 ~ ^ 發明內容 有鑑於此，本發明提供一種反饋式主動噪音控制電 路，其應用音訊補償電路來補償可能會被反饋式主動噪= 控制電路消除或改變之低頻音樂聲，並應用可分別調整S 益之加法器，以分別放大來自音訊補償電路之音訊補償^ 號，與來自帶通控制器之環境噪音訊號，降低或消除音^ 受到調整抗噪增益之影響。此外，更應用複數個麥克風$ 測器、電源延遲電路及開關，來改善主動噪音控制電路之 性能。 爲達上述及其他目的，本發明提供一種反饋式主動噪 200425766 1 1218twf.doc/006 音控制電路，此反饋式主動噪音控制電路包括：帶通控制 器、音訊補償電路、加法器及電流轉換增益器。 其中，帶通控制器用以接收麥克風感測器感測環境噪 音所得之噪音感測訊號，並調控噪音感測訊號之頻譜的增 益及相位’以產生環境噪音訊號。 音訊補償電路用以接收音樂發聲裝置所產生之音訊 輸入訊號，並產生高頻衰減較低頻爲大之音訊補償訊號， 以預爲補償可能會被此反饋式主動噪音控制電路消除或改 變之低頻音樂聲。 加法器具有可分別調整增益之第一輸入端及第二輸 入端，第一輸入端耦接帶通控制器，用以接收上述之環境 噪音訊號，予以適當處理成爲噪音消除訊號，驅動喇叭以 產生與環境噪音相位相反之聲波訊號，來抵消或降低低頻 環境噪音。而第二輸入端耦接音訊補償電路，用以接收上 述之音訊補償訊號’並將音訊補償訊號放大爲音訊輸出訊 號，以供喇叭產生使用者欲聆聽之音樂聲。 電流轉換增益器則用以接收噪音消除訊號與音訊輸 出訊號之合成訊號，並將其轉換爲電流訊號以驅動喇叭。 本發明之較佳實施例中，此反饋式主動噪音控制電路 更包括一電源延遲電路，此電源延遲電路用以接收供應此 〜反饋式主動噪音控制電路之電源，並於電源導通時，延 遲一預定時間，才將其電源供應至電流轉換增益器，以改 善反饋式主動噪音控制電路電源打開時所發出的異聲。 上述之電源延遲電路包括:延遲電路及電晶體。延遲電 200425766 112 1 8twf.doc/006 路例如是由串聯之一電阻與電容所組成，用以當電源導通 時，產生一延遲控制訊號。而電晶體具有一集極、一射極 及一基極，其中之基極耦接上述延遲電路，用以接收其延 遲控制訊號，並依據延遲控制訊號，來將集極接收之電源， 延遲一*預定時間導通至射極輸出。 此外，此反饋式主動噪音控制電路更包括一切換開 關，用以控制供應此反饋式主動噪音控制電路之電源，且 當切斷其電源時，並將音樂發聲裝置所產生之音訊輸入訊 號直接導通至喇叭輸出，以便當此反饋式主動噪音控制電 路之電源切斷時，仍可使用耳機來播放音樂，且音質不變。 在一實施例中，此反饋式主動噪音控制電路之音訊補 償電路包括:分別具有第一端及第二端之第一電阻、第二電 阻、第一電容、第二電容及第三電阻。其中，第一電阻之 第一端接收音樂發聲裝置所產生之音訊輸入訊號，並經第 一電容之第二端輸出其音訊補償訊號。其耦接關係則爲第 一電阻之第二端接地，第二電阻之第一端耦接第一電阻之 第一端，第一電容之第一端耦接第二電阻之第二端，第二 電容之第一端耦接第一電容之第二端，第三電阻之第一端 耦接第二電容之第二端，第三電阻之第二端則接地。 在一實施例中，上述之噪音感測訊號係由並聯連接之 複數個麥克風感測器，感測一環境噪音而得。而此反饋式 主動噪音控制電路，依據噪音感測訊號所產生之噪音消除 訊號，則輸出至喇叭，以產生與感測之環境噪音相位相反 之聲波訊號，來抵消或降低低頻環境噪音。 200425766 112 1 8twf. doc/006 爲讓本發明之上述和其他目的、特徵、和優點能更明 顯易懂，下文特以較佳實施例，並配合所附圖式，作詳細 說明如下： 實施方式： 請參考第1圖所示，其爲一般反饋式主動噪音控制耳 機之喇叭近場效應示意圖。圖中顯示，當裝置於喇叭110前 方之麥克風感測器120收集外界環境之白噪音（white noise) 時’因耳罩140腔體對於環境噪音之帶通特性，導致麥克風 感測器120會收集到低頻之環境噪音（約50Hz〜ΙΚΗζ)，居於 低頻之環境噪音頻率低、波長長之特性，因而對於麥克風 感測器120之裝置位置並無特別挑剔之處。 然而，當麥克風感測器120將收集到之低頻環境噪音 轉換爲噪音感測訊號，並傳送至主動噪音控制電路130，而 主動噪音控制電路130則依據此噪音感測訊號，來產生噪音 消除訊號，並將此噪音消除訊號傳送至喇叭110，以產生與 環境噪音相位相反之聲波訊號，來降低或消除麥克風感消fJ 器120所感測之低頻環境噪音時，由於喇叭11〇前方會医j胃 近場效應而產生音能渦流15〇，且因麥克風感測器12〇係放 置在靠近喇叭110前方附近區域，正好位於喇叭110前方所： 產生的近場效應的音能渦流150內。因此，麥克風感測器12〇 便會因爲近場效應的現象，而無法隨時淸晰、準確±也__ 低頻環境噪音，供送進主動噪音控制電路130，以產&、準_ 有效的反相聲波去抵消低頻噪音。 + 請參看第2圖所示，其爲喇叭位於自由音場之@ % 200425766 112 1 8twf.doc/006 應測量圖示。圖中應用一白噪聲產生器2 1 〇來模擬產生白噪 音訊號，再將此白噪音訊號傳送至喇叭220，以產生均勻穩 定的白噪音，然後使用音量計230在距離喇叭220不同的長 度L與角度α下量測其音壓位準，例如量測圖中具有相同長 度L，而角度相差α之A、Β兩點。由上述實驗中可知，在 距離喇叭220前方大於喇叭直徑之距離時，不同角度會量到 相同且穩定的音壓位準，但在距離喇叭220前方5〜10mm範 圍時，不同角度會量到完全不同且不穩定的音壓位準，驗 證了喇叭220前方近端之近場效應的存在及其影響。 請參看第3A與3B圖所示，其係分別顯示根據本發明較 佳實施例之喇叭前方裝置2與3個麥克風感測器之感測器安 裝位置示意圖。圖中顯示，本發明針對此一問題，在第3A 圖中靠近喇叭360前方附近區域放置了 310、320等2個麥克 風感測器，或在第3B圖中靠近喇叭37〇前方附近區域放置了 330、340及350等3個麥克風感測器，麥克風感測器310、 320、330、340及350的收音方向係指向喇叭前方中心線， 所以多個麥克風感測器310與320或330、340與350之間就會 因所在位置不同而收到不同淸晰度的訊號，如此截長補 短，藉以提高310與320或330、340與350等整組麥克風感測 器的收音品質，而使主動噪音控制電路產生準確、有效的 反相聲波去抵消低頻噪音，進而提高主動噪音控制耳機的 降噪性能。 請參看第4圖所示，其爲根據本發明較佳實施例之喇 叭前方裝置2個麥克風感測器之結構示意圖。圖中顯示，此 200425766 112 1 8twf. doc/006 反饋式主動噪音控制耳機400包括:2個麥克風感測器41〇與 420、主動噪音控制電路430及喇叭440。其中，使用裝置於 喇口八440前方周圍之2個麥克風感測器410與420來感測環境 噪音，並將環境噪音轉換爲一噪音感測訊號傳送至主動噪 音控制電路430，主動噪音控制電路430則依據接收之噪音 感測訊號，產生噪音消除訊號，以供喇卩八440產生與環境噪 音相位相反之聲波訊號，來降低或抵消低頻環境噪音。此 種作法之優點已如上述，因2個麥克風感測器410與420之所 在位置不同，會收到不同淸晰度的訊號，因而截長補短， 藉以提高410與420等整組麥克風感測器的收音品質，而使 主動噪音控制電路430產生準確、有效的反相聲波去抵消低 頻噪音，進而提高主動噪音控制耳機的降噪性能。 請參看第5圖所示，其爲根據本發明較佳實施例之一 種反饋式主動噪音控制電路方塊圖。圖中顯示，此反饋式 主動噪音控制電路500包括:帶通控制器510、音訊補償電路 520、加法器80、電流轉換增益器70及電源與開關電路530。 其中，帶通控制器510用以接收並聯連接之複數個麥 克風感測器51與52感測環境噪音所得之噪音感測訊號 SNI ’並調控噪音感測訊號sni之增益與相位，產生環境噪 音訊號SNO輸出至加法器80之第一輸入端801，以將環境噪 音訊號SNO放大爲噪音消除訊號，再經電流轉換增益器70 轉換爲電流訊號，並經RB/GR傳輸線以驅動喇叭，使喇叭 可產生與感測之環境噪音相位相反之聲波訊號，來消除或 降低其環境噪音。 "006 "006200425766 :1 8tvvf ( 加法器8〇之第二輸入端8〇2，則接收音樂發聲裝置（未 繪示）所產生之音訊輸入訊號LIN，以使喇叭可發出使用者 欲耳令聽之音樂聲，且此加法器80之第二輸入端802與前述之 第一輸入端8〇1的增益係可分別調整，以便爲了抗噪效果而 調整環境噪音訊號SNO之增益時，不會影響到音樂音量之 大小。 然而，因複數個麥克風感測器51與52，感測環境噪音 所得之噪音感測訊號SNI，通常也會包括使用者欲聆聽之音 樂聲，以致100Hz〜ΙΚΗζ之音樂聲，會隨著噪音訊號被部分 消除。爲了不致因此反饋式主動噪音控制電路500作用後， 讓使用者感覺到音樂聲之變化，而仍可聆聽到音質不變之 音樂，於是乃於音訊輸入訊號LIN輸入加法器80之第二輸入 端802前，插入一音訊補償電路520，以預先補償此可能被 部分消除之音樂聲。其補償方法爲先接收音樂發聲裝置所 產生之音訊輸入訊號LIN，然後經音訊補償電路520產生高 頻衰減較低頻爲大之音訊補償訊號LC，以預爲補償低頻音 樂聲之衰減，然後再輸入加法器80之第二輸入端802。 其中，音訊補償電路520如第6圖所示地包括:分別具有 第一端 8U、82卜 85 卜 84卜 831 及第二端 812、822、852、 842、832之第一電阻81、第二電阻82、第一電容85、第二 電容84及第三電阻83。第一電阻81之第一端811接收音樂發 聲裝虞所產生之音訊輸入訊號LIN，並經第·電容85之第二 端852輸出其音訊補償訊_LC。其耦接關係則爲第一電阻81 之第二端812接地，第二電阻82之第一端821耦接第一電阻 200425766 1 1218twf.doc/006 81之第一端811，第一電容85之第一端851耦接第二電阻82 之第二端822，第二電容84之第一端841耦接第一電容85之 第二端852，第三電阻83之第一端831耦接第二電容84之第 二端842，第三電阻83之第二端832則接地。 故知，第5圖之反饋式主動噪音控制電路500除了可以 補償部分可能被消除之音樂聲外，更因音訊輸入訊號LIN與 噪音感測訊號SNI之增益可以分別調整，使得音訊輸入訊號 LIN之增益不會受到調整抗噪增益之影響’而有穩定之音樂 音量，且不致產生低頻破音之情形。 此外，爲了改善當電源導通之瞬間’因電路尙未處於 穩定狀態，而可能自喇叭發出瞬間異音之情形。故如第5圖 所示，此反饋式主動噪音控制電路500更包括一電源與開關 電路530，在電源與開關電路530中具有如第7圖所示之電源 延遲電路540，以接收供應此反饋式主動噪音控制電路500 之電源BATT，並於電源V+導通時，延遲一預定時間’才將 其電源POW供應至電流轉換增益器70 ’其工作原理說明如 下。 如第7圖所示，此電源延遲電路540包括電晶體90及由 串聯之電阻91與電容92所組成的延遲電路560。電晶體90具 有一集極9〇1、一射極903及一基極902，電阻91具有第一端 911及第二端912，電容92具有第一端921及第二端922。其 中之集極901連接電池97供應之電源BATT，基極902連接電 阻91之第二端912與電容92之第一端921，電容92之第一端 922則接地。 200425766 1 1218twf.doc/006 當電阻91之第一端911所連接之電源V+導通時（V+供 應給其他電路所需電源），將經由電容92充電，以產生一延 遲控制訊號。此延遲控制訊號將使得電晶體90延遲一預定 時間後才導通，以致由電晶體90之射極903輸出之電源 POW，將延遲供應至第5圖中之電流轉換增益器70。故當電 源V+導通時，此反饋式主動噪音控制電路500便不會自喇叭 98發出瞬間之異音。 如第7圖所示，此電源與開關電路530另包括一切換開 關550。此切換開關550除了用以控制供應此反饋式主動噪 音控制電路500之電源V+的導通與否外，更於切換開關550 切斷電源V+時，將音樂發聲裝置（未繪示）所產生之音訊輸 入訊號LIN直接導通至喇叭98,以便當此反饋式主動噪音控 制電路500之電源V+未導通時，仍可使用耳機來播放音樂， 且音質不變。 以下將以一實驗來證明使用複數個麥克風感測器之 反饋式主動噪音控制耳機的降噪效果。其進行方式爲將第4 圖之反饋式主動噪音控制耳機400戴在一模擬人工頭上，而 此模擬人工頭具有測量其收聽到之不同頻率的噪音音量之 功能，以便分別測量並記錄其結果如表一及表二所示。表 一爲僅接通麥克風感測器410至主動噪音控制電路430時之 測量結果，表二則爲同時接通麥克風感測器410與420至主 動噪音控制電路430時之測量結果，其中之ANC-OFF欄位爲 關閉主動噪音控制電路430運作時之測量値，亦即未消除 200425766 1 1218twf.doc/006 頻率 (Hz) ANC-OFF 噪音量（分貝） ANC-0N 噪音量（分貝） 降噪量 50 -44.892052 -46.355553 1.463501 63 -47.250725 -51.611275 4.36055 80 -46.059258 -52.916901 6.857643 100 -39.596458 -50.056454 10.46 125 -40.698879 -52.588493 11.88961 160 -44.13002 -57.5037 13.37368 200 -49.081154 -58.509605 9.428451 250 -51.771255 -60.032673 8.261418 315 -59.943424 -69.942879 9.999455 400 -68.614731 -79.727463 11.11273 500 -72.215195 -83.750633 1 1.53544 630 -73.721779 -82.608246 8.886467 800 -72.781471 -79.317261 6.53579 1000 -79.337273 -76.014885 -3.32239 平均 9.42467754 表一 200425766 1 1218twf.doc/006 頻率 (Hz) ANC-0FF 噪音量（分貝） ANC-0N 噪音量（分貝） 降噪量 50 -59.613277 -61.625042 2.011765 63 -59.073704 -64.525406 5.451702 80 -54.281155 -61.29026 7.009105 100 -47.093666 -57.906025 10.81236 125 -42.541756 -57.877411 15.33566 160 -44.581146 -62.431255 17.85011 200 -42.310223 -59.130478 16.82026 250 -51.757565 -63.474697 11.71713 315 -57.003044 -68.348465 11.34542 400 -63.156078 -75.24823 12.09215 500 -63.727406 -79.12429 15.39688 630 -71,959145 -83.755692 11.79655 800 -69.567673 -76.19136 6.623687 1000 -77.687004 -73.204292 -4.48271 平均 12.7675294 噪音時，模擬人工頭收聽到之噪音音量，而anc-on欄位 在表一中爲僅接通麥克風感測器410至主動噪音控制電路 430運作時，模擬人工頭收聽到之噪音音量，在表二中則爲 同時接通麥克風感測器410與420至主動噪音控制電路430 17 200425766 1121 8twf.doc/〇〇6 運作時，模擬人工頭收聽到之噪音音量。在表一及表二之 最後一列則分別爲其平均降噪量。參考其平均降噪量，在 僅接通麥克風感測器410至主動噪音控制電路430時之平均 降噪量爲9.42467754分貝，而在同時接通麥克風感測器 410與420至主動噪音控制電路430時之平均降噪量爲 12.7675294分貝。故知，使用兩個麥克風感測器的設計，再 配合本發明之反饋式主動噪音控制電路500，確實可獲得較 佳之降噪效果。 雖然本發明已以較佳實施例揭露如上，然其並非用以 限定本發明，任何熟習此技藝者，在不脫離本發明之精神 和範圍內，當可作各種之更動與潤飾，因此本發明之保護 範圍當視後附之申請專利範圍所界定者爲準。 圖式簡單說明： 第1圖係顯示一般反饋式主動噪音控制耳機之喇叭近 場效應示意圖； 第2圖係顯示喇叭位於自由音場之近場效應測量圖； 第3A及3B圖係分別顯示根據本發明較佳實施例之喇 叭前方裝置2與3個麥克風感測器之感測器安裝位置示意 圖； 第4圖係顯示根據本發明較佳實施例之喇叭前方裝置 2個麥克風感測器之結構示意圖； 第5圖係顯示根據本發明較佳實施例之一種反饋式主 動噪音控制電路方塊圖； 第6圖係顯示根據本發明較佳實施例之一種反饋式主 200425766 1 1218twf.doc/006 動噪音控制電路的音訊補償電路圖；以及 第7圖係顯示根據本發明較佳實施例之一種反饋式主 動噪音控制電路的電源與開關電路圖。 圖式標示說明= 70電流轉換增益器 81、82、83、91 電阻 84、85、92 電容 80加法器 90電晶體 97電池 98、110、220、360、370、440 喇叭 51、52、120、310〜350、410、420 麥克風感測器 130、430主動噪音控制電路 140耳罩 15 0音能渦流 210白噪聲產生器 230音量計 400反饋式主動噪音控制耳機 500反饋式主動噪音控制電路 510帶通控制器 520音訊補償電路 530電源與開關電路 540電源延遲電路 550切換開關 200425766 1 1218twf.doc/006 560延遲電路 811、821、831、841、851、911、921 第一端 821、822、832、842、852、912、922 第二端 801第一輸入端 802第二輸入端 901集極 902基極 903射極200425766 1 12l8twf. Doc / 006 TECHNICAL FIELD The present invention relates to a headset, and more particularly to a feedback-type active noise control circuit for a feedback-type active noise control headset. Pre-i technology Today, with the increasing popularity of electrified products, audio equipment has become an important device for people to entertain and acquire new knowledge. Among them, earphones are convenient devices for listening anytime, anywhere. In order to provide better listening results, noise prevention methods must be adopted for the ambient noise that the earphone users will listen to simultaneously. According to the different methods of noise prevention, headphones can be divided into two types: passive noise prevention headphones and active noise control headphones. Passive dullness, because anti-earphones simply rely on sound insulation materials to reduce environmental noise, so the ability to resist noise is greatly related to the material, thickness, structural design, fit, etc. of the materials used. These headphones are generally very large and heavy. In addition, due to the material used in passive noise prevention headphones, there is almost no ability to block low-frequency noise, so that low-frequency noise such as engines and blowers has almost no noise prevention effect. Conversely, active noise control headsets do not have the above restrictions, and are therefore very popular with consumers. However, commercially available active noise control headphones usually only have one microphone sensor in front of each of their left and right speakers. This method, no = where it is placed, can only rely on a microphone sensor to receive the noise signal in front of the speaker, so that the performance requirements of the microphone sensor itself are very high, in addition to selecting high sensitivity and expensive Microphone = outside of the tester ', even during production and assembly, the microphone's sensing is more ensured ^ 20042004766 1 1 2 1 8tvvf.doc / 006 can maintain the original high sensitivity, making the welding process very vulnerable during mass production And microphone sensors, which affects the yield and cost of production. Furthermore, the microphone sensors of general feedback-type active noise control headphones are placed in a distance of 0 ^ ~ lcm in front of the speaker, which will produce serious near-field effects, so even if a high-sensitivity microphone sensor is used, it will still be affected. The effect of the near-field effect in front of the horn causes the noise reduction effect to be greatly reduced. In addition, the active noise control circuit in the conventional feedback-type active noise control headset fails to consider the noise input obtained from the audio input signal generated by a music sounding device such as a player and the microphone sensor to sense ambient noise The δΛ 5 tiger's gain adjustment circuit is separated, which causes the noise sensing signal's gain to be adjusted for anti-noise effect, which will affect the original frequency spectrum of the music, which may cause low-frequency breaking, or open the active noise control circuit when the user listens to music When the power is turned on, the volume of the music will suddenly increase, resulting in extremely comfortable ears. ~ ^ SUMMARY OF THE INVENTION In view of this, the present invention provides a feedback-type active noise control circuit, which employs an audio compensation circuit to compensate for low-frequency music sounds that may be eliminated or changed by the feedback-type active noise = control circuit, and can be adjusted separately by S The effective adder is to amplify the audio compensation signal from the audio compensation circuit and the environmental noise signal from the band-pass controller, respectively, to reduce or eliminate the noise ^ affected by adjusting the anti-noise gain. In addition, multiple microphone detectors, power delay circuits, and switches are used to improve the performance of the active noise control circuit. To achieve the above and other objectives, the present invention provides a feedback active noise control circuit 200425766 1 1218twf.doc / 006. The feedback active noise control circuit includes: a band-pass controller, an audio compensation circuit, an adder, and a current conversion gain. Device. Among them, the band-pass controller is used to receive the noise sensing signal obtained by the microphone sensor sensing the environmental noise, and adjust the gain and phase of the frequency spectrum of the noise sensing signal to generate the environmental noise signal. The audio compensation circuit is used to receive the audio input signal generated by the music sounding device, and generate the audio compensation signal with high frequency attenuation and lower frequency. The low frequency is expected to compensate for the low frequency that may be eliminated or changed by this feedback-type active noise control circuit. The sound of music. The adder has a first input terminal and a second input terminal that can adjust the gain, respectively. The first input terminal is coupled to a band-pass controller to receive the above-mentioned environmental noise signal, which is appropriately processed into a noise cancellation signal, and drives the speaker to generate A sound wave signal with a phase opposite to the ambient noise to cancel or reduce low-frequency ambient noise. The second input terminal is coupled to the audio compensation circuit, and is used for receiving the above-mentioned audio compensation signal 'and amplifying the audio compensation signal into an audio output signal for the speaker to generate the music sound that the user wants to listen to. The current conversion gain is used to receive the combined signal of noise cancellation signal and audio output signal, and convert it into a current signal to drive the speaker. In a preferred embodiment of the present invention, the feedback-type active noise control circuit further includes a power delay circuit. The power-delay circuit is used to receive the power supplied to the ~ feedback-type active noise control circuit, and delays a time when the power is on. The power is supplied to the current conversion gain device at a predetermined time to improve the abnormal sound produced when the power of the feedback-type active noise control circuit is turned on. The power supply delay circuit includes a delay circuit and a transistor. Delay circuit 200425766 112 1 8twf.doc / 006 The circuit is composed of a resistor and a capacitor in series, for example, to generate a delay control signal when the power is turned on. The transistor has a collector, an emitter, and a base. The base is coupled to the delay circuit to receive the delay control signal, and according to the delay control signal, the power received by the collector is delayed by one. * Conducted to emitter output for a predetermined time. In addition, the feedback active noise control circuit further includes a switch for controlling the power supply to the feedback active noise control circuit, and when the power is cut off, the audio input signal generated by the music sound device is directly turned on. To the speaker output so that when the power of this feedback-type active noise control circuit is cut off, you can still use headphones to play music with the same sound quality. In one embodiment, the audio compensation circuit of the feedback-type active noise control circuit includes a first resistor, a second resistor, a first capacitor, a second capacitor, and a third resistor, respectively, having a first terminal and a second terminal. The first terminal of the first resistor receives the audio input signal generated by the music sounding device, and outputs its audio compensation signal through the second terminal of the first capacitor. The coupling relationship is that the second terminal of the first resistor is grounded, the first terminal of the second resistor is coupled to the first terminal of the first resistor, the first terminal of the first capacitor is coupled to the second terminal of the second resistor, and the first The first terminal of the two capacitors is coupled to the second terminal of the first capacitor, the first terminal of the third resistor is coupled to the second terminal of the second capacitor, and the second terminal of the third resistor is grounded. In one embodiment, the above-mentioned noise sensing signal is obtained by sensing a plurality of ambient noises by a plurality of microphone sensors connected in parallel. This feedback-type active noise control circuit outputs the noise cancellation signal generated by the noise sensing signal to the speaker to generate a sound wave signal with a phase opposite to the sensed ambient noise to cancel or reduce low-frequency ambient noise. 200425766 112 1 8twf. Doc / 006 In order to make the above and other objects, features, and advantages of the present invention more comprehensible, the following is a detailed description with preferred embodiments and the accompanying drawings as follows: Embodiment : Please refer to Figure 1, which is a schematic diagram of the near field effect of a general feedback active noise control headset. The figure shows that when the microphone sensor 120 installed in front of the speaker 110 collects white noise from the external environment, the microphone sensor 120 will collect Ambient noise to low frequencies (about 50Hz to 1KΗζ), which is characterized by low frequencies and long wavelengths, is not particularly critical of the device position of the microphone sensor 120. However, when the microphone sensor 120 converts the collected low-frequency ambient noise into a noise sensing signal and transmits it to the active noise control circuit 130, the active noise control circuit 130 generates a noise cancellation signal based on the noise sensing signal. And transmit this noise canceling signal to the speaker 110 to generate a sound wave signal with a phase opposite to the ambient noise to reduce or eliminate the low frequency environmental noise sensed by the microphone canceling fJ 120, because the front of the speaker 11 will heal the stomach The near field effect generates a sound energy eddy current of 150, and the microphone sensor 12 is placed near the front of the speaker 110, and is located right in front of the speaker 110: The near field effect generated by the sound energy eddy current 150. Therefore, the microphone sensor 120 will not be clear and accurate at any time because of the near-field effect. It is also __ low-frequency ambient noise, which is fed into the active noise control circuit 130 to produce & Inverted sound waves to cancel low frequency noise. + Please refer to Figure 2, which is a diagram of the measured @@ 20042004766 112 1 8twf.doc / 006 where the speaker is located in the free sound field. In the figure, a white noise generator 2 1 0 is used to simulate the generation of a white noise signal, and this white noise signal is transmitted to the speaker 220 to generate a uniform and stable white noise. Then, the volume meter 230 is used at a different length L from the speaker 220. Measure the sound pressure level with the angle α, for example, the measurement chart has the same length L, and the angles differ by A and B points α. It can be known from the above experiments that when the distance from the front of the speaker 220 is larger than the diameter of the speaker, different angles will measure the same and stable sound pressure level, but when the distance is 5 ~ 10mm from the front of the speaker 220, the different angles will be measured completely. Different and unstable sound pressure levels verify the existence and impact of near-field effects near the front of the horn 220. Please refer to Figs. 3A and 3B, which are schematic diagrams showing the installation positions of the sensors in front of the horn 2 and the three microphone sensors according to the preferred embodiment of the present invention, respectively. As shown in the figure, the present invention addresses this problem. In FIG. 3A, two microphone sensors such as 310 and 320 are placed near the area in front of the speaker 360, or in the area near the front of the speaker 37 in FIG. 3B. Three microphone sensors, 330, 340, and 350. The microphone receivers 310, 320, 330, 340, and 350 receive the sound from the center line in front of the speaker. Therefore, multiple microphone sensors 310 and 320 or 330 or 340 Between 350 and 350, signals with different degrees of clarity will be received depending on their location. Such truncation and supplementation will improve the radio quality of the entire group of microphone sensors, such as 310 and 320 or 330, 340, and 350. The active noise control circuit generates accurate and effective inverse sound waves to cancel low-frequency noise, thereby improving the noise reduction performance of active noise control headphones. Please refer to FIG. 4, which is a schematic structural diagram of two microphone sensors of a front device of a speaker according to a preferred embodiment of the present invention. As shown in the figure, this 200425766 112 1 8twf.doc / 006 feedback active noise control headset 400 includes: two microphone sensors 41 and 420, an active noise control circuit 430, and a speaker 440. Among them, two microphone sensors 410 and 420 installed around the front of Lakou Ba 440 are used to sense the ambient noise, and the ambient noise is converted into a noise sensing signal and transmitted to the active noise control circuit 430, the active noise control circuit 430 generates a noise cancellation signal based on the received noise sensing signal, so that Lao Ba 440 generates a sound wave signal with a phase opposite to the ambient noise to reduce or cancel the low-frequency ambient noise. The advantages of this method are as described above. Because the two microphone sensors 410 and 420 are located differently, they will receive signals with different degrees of clarity. The sound quality of the measurement device causes the active noise control circuit 430 to generate accurate and effective inverse sound waves to cancel low-frequency noise, thereby improving the noise reduction performance of the active noise control headset. Please refer to FIG. 5, which is a block diagram of a feedback type active noise control circuit according to a preferred embodiment of the present invention. As shown in the figure, the feedback-type active noise control circuit 500 includes: a band-pass controller 510, an audio compensation circuit 520, an adder 80, a current conversion gain device 70, and a power supply and switch circuit 530. Among them, the band-pass controller 510 is used to receive the noise sensing signal SNI 'obtained by sensing the environmental noise of a plurality of microphone sensors 51 and 52 connected in parallel, and to adjust the gain and phase of the noise sensing signal sni to generate an environmental noise signal SNO is output to the first input terminal 801 of the adder 80 to amplify the environmental noise signal SNO into a noise cancellation signal, and then convert it into a current signal through the current conversion gainer 70, and drive the speaker through the RB / GR transmission line, so that the speaker can be used. Generates a sound wave signal in the opposite phase to the sensed ambient noise to eliminate or reduce its ambient noise. " 006 " 006200425766: 1 8tvvf (the second input terminal 802 of the adder 80, receives the audio input signal LIN generated by the music sound device (not shown), so that the speaker can emit the user's ear Make the listening music sound, and the gain of the second input terminal 802 of the adder 80 and the aforementioned first input terminal 801 can be adjusted separately, in order to adjust the gain of the environmental noise signal SNO for the anti-noise effect. It will affect the volume of the music volume. However, the noise sensing signal SNI obtained by sensing the ambient noise due to the plurality of microphone sensors 51 and 52 usually also includes the music sound that the user wants to listen to, so that the frequency of 100Hz ~ ΙΚΗζ The music sound will be partially eliminated with the noise signal. In order not to cause the feedback-type active noise control circuit 500 to function, the user can feel the change of the music sound and still hear the music with the same sound quality. An audio signal compensation circuit 520 is inserted in front of the second input terminal 802 of the input signal LIN input adder 80 to pre-compensate this music sound that may be partially eliminated. The compensation method is to connect first The audio input signal LIN generated by the sound receiving device is then used by the audio compensation circuit 520 to generate a high-frequency attenuation low-frequency audio compensation signal LC to pre-compensate the attenuation of low-frequency music sound, and then input to the adder 80. The second input terminal 802. As shown in FIG. 6, the audio compensation circuit 520 includes: a first terminal 8U, 82, 85, 84, 831, and a second terminal 812, 822, 852, 842, 832, respectively. A resistor 81, a second resistor 82, a first capacitor 85, a second capacitor 84, and a third resistor 83. The first terminal 811 of the first resistor 81 receives the audio input signal LIN generated by the music sounding device, and passes the first · The second terminal 852 of the capacitor 85 outputs its audio compensation signal _LC. The coupling relationship is that the second terminal 812 of the first resistor 81 is grounded, and the first terminal 821 of the second resistor 82 is coupled to the first resistor 200425766 1 1218twf. The first terminal 811 of doc / 006 81, the first terminal 851 of the first capacitor 85 is coupled to the second terminal 822 of the second resistor 82, and the first terminal 841 of the second capacitor 84 is coupled to the second terminal of the first capacitor 85 852, the first terminal 831 of the third resistor 83 is coupled to the second terminal 842 of the second capacitor 84, the third The second end 832 of the resistor 83 is grounded. Therefore, the feedback active noise control circuit 500 in FIG. 5 can compensate for the part of the music sound that may be eliminated, and also because of the gain of the audio input signal LIN and the noise sensing signal SNI. Adjust separately so that the gain of the audio input signal LIN will not be affected by the adjustment of the anti-noise gain, and there will be a stable music volume and no low-frequency cracking. In addition, to improve the moment when the power is turned on, the circuit is not It is in a stable state, and an abnormal sound may be emitted from the speaker. Therefore, as shown in FIG. 5, the feedback active noise control circuit 500 further includes a power supply and a switching circuit 530. The power supply and the switching circuit 530 has a power supply delay circuit 540 as shown in FIG. 7 to receive and supply the feedback. The power source BATT of the active noise control circuit 500 is delayed for a predetermined time when the power source V + is turned on, and then the power source POW is supplied to the current conversion gain unit 70. Its working principle is described below. As shown in FIG. 7, the power supply delay circuit 540 includes a transistor 90 and a delay circuit 560 composed of a resistor 91 and a capacitor 92 connected in series. The transistor 90 has a collector 901, an emitter 903, and a base 902. The resistor 91 has a first terminal 911 and a second terminal 912, and the capacitor 92 has a first terminal 921 and a second terminal 922. The collector 901 is connected to the power supply BATT supplied by the battery 97, the base 902 is connected to the second terminal 912 of the resistor 91 and the first terminal 921 of the capacitor 92, and the first terminal 922 of the capacitor 92 is grounded. 200425766 1 1218twf.doc / 006 When the power source V + connected to the first terminal 911 of the resistor 91 is turned on (V + is supplied to the power required by other circuits), it will be charged through the capacitor 92 to generate a delay control signal. This delay control signal will cause the transistor 90 to turn on after a predetermined time delay, so that the power supply POW output from the emitter 903 of the transistor 90 will supply the delay to the current conversion gain device 70 in FIG. 5. Therefore, when the power source V + is turned on, the feedback-type active noise control circuit 500 will not emit an instantaneous noise from the horn 98. As shown in FIG. 7, the power supply and switching circuit 530 further includes a switching switch 550. The switch 550 is used to control whether the power supply V + of the feedback-type active noise control circuit 500 is turned on or not, and the switch 550 cuts off the power V + to generate audio generated by a music sound device (not shown). The input signal LIN is directly connected to the speaker 98, so that when the power supply V + of the feedback active noise control circuit 500 is not connected, the headphones can still be used to play music with the same sound quality. In the following, an experiment will be used to demonstrate the noise reduction effect of feedback active noise control headphones using multiple microphone sensors. The method is to put the feedback active noise control earphone 400 of FIG. 4 on an artificial artificial head, and the artificial artificial head has a function of measuring the noise volume of different frequencies it listens to, so as to separately measure and record the results such as Tables 1 and 2 show. Table 1 is the measurement results when only the microphone sensor 410 is connected to the active noise control circuit 430, and Table 2 is the measurement results when the microphone sensors 410 and 420 are connected to the active noise control circuit 430 at the same time, of which ANC The -OFF field is the measurement when the active noise control circuit 430 is turned off, that is, 200425766 1 1218twf.doc / 006 frequency (Hz) ANC-OFF noise level (dB) ANC-0N noise level (dB) Noise reduction Volume 50 -44.892052 -46.355553 1.463501 63 -47.250725 -51.611275 4.36055 80 -46.059258 -52.916901 6.857643 100 -39.596458 -50.056454 10.46 125 -40.698879 -52.588493 11.88961 160 -44.13002 -57.5037 13.37368 200 -49.081154-58.509.503 5.589.509605. -59.943424 -69.942879 9.999455 400 -68.614731 -79.727463 11.11273 500 -72.215195 -83.750633 1 1.53544 630 -73.721779 -82.608246 8.886467 800 -72.781471 -79.317261 6.53579 1000 -79.337273 -76.014885 -3.32239 Average 9.42467754 Table one 200425766 1 Hz) ANC-0FF Noise level (dB) ANC- 0N Noise amount (dB) Noise reduction amount 50 -59.613277 -61.625042 2.011765 63 -59.073704 -64.525406 5.451702 80 -54.281155 -61.29026 7.009105 100 -47.093666 -57.906025 10.81236 125 -42.541756 -57.877411 15.33566 160 -44.581146 -62.431255 17.850 11.850 16.82026 250 -51.757565 -63.474697 11.71713 315 -57.003044 -68.348465 11.34542 400 -63.156078 -75.24823 12.09215 500 -63.727406 -79.12429 15.39688 630 -71,959145 -83.755692 11.79655 800 -69.567673 -76.19136 6.623687 1000 -77.687294 -73 -67.687294 -73 , Simulate the volume of the noise heard by the artificial head, and the anc-on field in Table 1 is only when the microphone sensor 410 is connected to the active noise control circuit 430, and the noise level of the artificial head is simulated, as shown in Table 2. In the middle, the microphone sensors 410 and 420 are simultaneously turned on to the active noise control circuit 430 17 200425766 1121 8twf.doc / 〇〇6 When operating, it simulates the volume of noise heard by the artificial head. The last columns in Tables 1 and 2 are their average noise reductions, respectively. With reference to its average noise reduction, the average noise reduction when only the microphone sensor 410 is connected to the active noise control circuit 430 is 9.42467754 dB, while the microphone sensors 410 and 420 are simultaneously connected to the active noise control circuit 430 The average noise reduction is 12.7675294 dB. Therefore, it is known that the design using two microphone sensors and the feedback active noise control circuit 500 of the present invention can indeed achieve better noise reduction effects. Although the present invention has been disclosed as above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make various modifications and retouches without departing from the spirit and scope of the present invention. Therefore, the present invention The scope of protection shall be determined by the scope of the attached patent application. Brief description of the drawings: Figure 1 is a schematic diagram of the near field effect of a general feedback active noise control headset; Figure 2 is a measurement diagram of the near field effect of the speaker in a free sound field; Figures 3A and 3B are shown according to Schematic diagram of the sensor installation positions of the front horn device 2 and three microphone sensors of the preferred embodiment of the present invention; FIG. 4 shows the structure of the two microphone sensors of the front horn device according to the preferred embodiment of the present invention Figure 5 is a block diagram of a feedback active noise control circuit according to a preferred embodiment of the present invention; Figure 6 is a feedback master according to a preferred embodiment of the present invention 200425766 1 1218twf.doc / 006 An audio compensation circuit diagram of a noise control circuit; and FIG. 7 is a power supply and switching circuit diagram of a feedback active noise control circuit according to a preferred embodiment of the present invention. Graphical description = 70 current conversion gains 81, 82, 83, 91 resistors 84, 85, 92 capacitors 80 adders 90 transistors 97 batteries 98, 110, 220, 360, 370, 440 speakers 51, 52, 120, 310 ~ 350, 410, 420 microphone sensor 130, 430 active noise control circuit 140 earmuffs 15 0 acoustic energy eddy current 210 white noise generator 230 volume meter 400 feedback active noise control headset 500 feedback active noise control circuit 510 with Communication controller 520 Audio compensation circuit 530 Power supply and switching circuit 540 Power delay circuit 550 Switching switch 200425766 1 1218twf.doc / 006 560 Delay circuit 811, 821, 831, 841, 851, 911, 921 First end 821, 822, 832 , 842, 852, 912, 922 Second end 801 First input end 802 Second input end 901 Collector 902 Base 903 Emitter