200926138 九、發明說明 【發明所屬之技術領域】 本發明’係有關於適合於使用在如同雜音消除頭戴式 耳機或是耳機、耳機揚聲器等一般在耳朵之極近旁處設置 揚聲器的收聽機器中之雜音控制系統。 【先前技術】 〇 近年來’伴隨著攜帶式音樂播放器的受歡迎，各種之 雜音消除頭戴式耳機（耳機）係變得普及。雜音消除頭戴 式耳機，係在收聽點近旁處，設置噪音源資訊檢測感測器 或是控制狀態評價感測器，若是前者之情況，則藉由通過 對從感測器起直到收聽點爲止之傳達特性作了模擬的濾波 器，來使收聽點之噪音準位衰減；若是後者之情況，則構 成不會使嘷鳴（howling )發生之回歸濾波器，並藉由對 音場附加逆相之訊號，來使收聽點之噪音準位衰減。 © 現在於市面上所販賣之雜音消除頭戴式耳機的大半， 係爲類比方式之雜音消除頭戴式耳機。此方式，係藉由被 _ 組入至頭戴式耳機中之麥克風，來將周圍之聲音（相對於 再生音，由於此係爲噪音，因此，以下，係稱爲「噪音」 )作捕捉，並將所捕捉到之噪音的相位作反轉，再加算至 從播放器而來之再生訊號中。於此構造中，從外部而侵入 至頭戴式耳機內部之噪音，係藉由相位被作了反轉之訊號 而被抵消，而只有從播放器而來之再生訊號會進入至使用 者之耳朵中。 -4- 200926138 在最近’雖然亦提案有數位方式之雜音消除頭戴式耳 機’但是’一般而言’雜音消除頭戴式耳機，係成爲將感 測器與揚聲器之收聽點近接配置的構成，而多所會有轉換 器所致之延遲時間超過了音波之到達時間差的情況，而在 A/D變換以及D/A變換中係需要大的延遲時間，因此 ’現狀而言，數位訊號處理係幾乎未被作採用。 作爲類比方式雜音消除頭戴式耳機之先前技術，例如 0 ’係存在有專利文獻1或專利文獻2所記載之發明等。又 ，作爲數位方式雜音消除頭戴式耳機之先前技術，例如， 係存在有專利文獻3所記載之發明。 〔專利文獻1〕日本特開平11-308685號公報 〔專利文獻2〕日本特開平11-237889號公報 〔專利文獻3〕日本特開平09-93684號公報 【發明內容】 〇 〔發明所欲解決之課題〕 因此，在前述一般之先前技術中，係存在有下述一般 之問題點，並期望能解決此些問題。 (1)在噪音下之立體再生 在既存的以所請立體再生爲目的之雙耳系統中，在某 種程度以上之噪音下，會有由於掩蔽效應（masking effect )而使得音像無法作定位的問題點。在現今之廣泛 普及的雜音消除頭戴式耳機、雜音消除耳機中，對於音源 -5- 200926138 之修正濾波器，係藉由低次之類比濾波器而構成，並僅停 留在對衰減之帶域作強調一事，在要求有詳細之相位資訊 的雙耳系統中之使用，係仍有困難點。 (2 )注目於舒適性之適應化 既存之雜音消除頭戴式耳機、雜音消除耳機，或是與 其相關之硏究報告等，幾乎均係對於成爲其之控制對象的 〇 雜音之衰減量作注目者。然而，作爲現實之問題，於低噪 音化中，係週知有：若是進行過度之雜音消除，則係會伴 隨有壓迫感或是不適感，又，由於有時候在雜音消除電路 本身中，亦會產生有寄生於麥克風擴大器等之中的雜音， 因此，若是不論於何種環境噪音準位下均將雜音消除量設 定在極限，則係並不一定能夠達成舒適性之提昇。 亦即是，在先前之頭戴式耳機或是耳機所致之主動雜 音消除系統中，比例於噪音之壓抑量，有時候會感到有壓 €1 迫感或是逆相感。又，在多數之雜音消除頭戴式耳機中， 係會發現到：由於控制濾波部之濾波次數的限制等，特別 是在高頻等之壓抑對象帶域外的雜訊準位會上升的現象。 特別是，在先前之固定濾波器所致的主動雜音消除系統中 ，係可以發現到：特別是在低準位噪音狀況下的壓迫感、 或是雜音準位之上升等的問題。 (3 )對於參數限制所致之非週期音的適應處理 在廣泛普及之雜音消除頭戴式耳機、雜音消除耳機中 -6 - 200926138 ’其所採用之全類比系統，在構成上，對特性之更新係爲 困難。又，在學術領域或是導管（duct)等之其他構成的 ANC ( active noise can cal，主動雜音消除）中而被廣泛進 行有硏究的LMS (Least Mean Square)演算法等之使用有 數位濾波器的雜音消除系統’在如同頭戴式耳機一般之將 參考麥克風與揚聲器、誤差麥克風極端地近接配置的構成 中’由於A/D變換以及D/A變換所伴隨之延遲，而係 〇 被要求有高取樣’而’取樣頻率之大幅的上升，係會使計 算量變得膨大’而在非週期性之噪音控制中，多係無法實 用化。 (4 )對個人差之適應 在耳機型之雜音消除系統中，基本上，爲了在前授· 後授之兩手法中得到最適之結果，係有必要因應於在收聽 者各人之裝著狀態中所特有之從參考麥克風起直到揚聲器 ® 之傳達函數、從揚聲器起直到誤差麥克風的傳達函數，來 進行處理。但是’在先前技術中，係僅使用有系統側所預 先準備之傳達函數’並進行與其相對應之處理，因此，無 法進行對於每一個收聽者而作了適應化之處理。 (5 )對裝置差之適應 在既存之雜音消除頭戴式耳機·雜音消除耳機系統中 ’由於係爲在1個裝置中具有1個處理系之構成，因此， 係有必要分別準備具有各種之特性的雜音消除用之裝置， 200926138 而像是例如在1個機種之攜帶式音樂播放器中而連接各種 的雜音消除用裝置並作使用一事，係爲不可能。 本發明，係爲了解決如同上述一般之先前技術的問題 點而提案者，其目的，係在於提供一種：具備有在保持雜 音消除功能的同時亦能夠對係數作變換之特徵，並藉由此 ，而在更新項中，不僅是對象雜音之衰減量，且亦將低雜 音下之不適感組入至評價函數中，而實現將舒適性作爲評 © 價基準的雜音控制系統。 本發明之其他目的，係在於提案一種：將對於評價函 數之適應處理段的更新項限定爲數段，並藉由可作數位控 制之類比元件來構成雜音消除系統，藉由此，而在具備有 高速之回應性的同時，亦能夠以極爲現實之計算量來對參 數作更新，而發揮有類比•數位之優點的具備實用性之系 統。 特別是，本發明，係將針對在耳朵之近旁而使用的耳 〇 機、頭戴式耳機、耳機揚聲器等而能夠達成前述一般之目 的一事，作爲其中一種形態。其理由，係如下所述。 亦即是，對於各種之噪音場進行音場控制並創造出對 於人類而言係爲舒適的音環境之所謂的主動雜音控制，可 以說是受到以DSP爲代表之數位訊號處理的發展之影響 而使實用化急速的進行之技術的其中之一。例如，從在空 調等之導管處的雜音消除或是雜音消除頭戴式耳機起，直 到車內之雜音控制或是工廠或高速公路等之雜音消除，不 拘於規模之大小，硏究均係在進行。然而，在耳機、頭戴 -8 - 200926138 式耳機、耳機揚聲器處之雜音控制，在特殊之狀況下’係 可列舉出下述一般之特徵。 (η從所構成之系統規模來說’一般而言’參考麥 克風-外耳道間以及參考麥克風-揚聲器間之距離，係極 端的短，因此，由於在A/D變換或是D/A變換中所伴 隨之延遲，數位訊號處理所致之控制濾波器的構成，多係 有不利之處。 © ( 2 )由於控制對象點係爲鼓膜之近旁，因此，依存 於控制量，會有感受到壓迫感、或是對於由於控制所產生 之新的噪音成爲敏感等的特徵。若是採用廣泛地被嘗試之 僅藉由誤差麥克風所觀察到之雜音量的實效値來作評價的 方法，則會有此評價與人類所實際感受到的舒適性並無法 作等同的情況。 在本發明中，係對應於此些之（1) 、（2)，而提案 ® ( a )數位訊號，係僅使用在控制中，且信號係僅經 過類比路徑，而對在耳機等之中的雜音消除作了特化之發 揮有數位•類比之優點的數位控制·類比雜音消除系統。 (b)並不僅將雜音功率之控制量作爲評價基準，而 亦根據包含有主觀評價之注目於聽者之舒適性的評價函數 ，來進行適應化。 〔用以解決課題之手段〕 爲了達成上述目的’本發明之雜音控制系統，其特徵 -9- 200926138 ，係爲使用電子音量器（volume)或是電子较減器（ attenuator )等之可變元件，來將構成前授處理部或是後 授處理部等之雜音消除處理部的濾波器（頻率回應控制部 )與增益放大器（增益控制部）從外部而控制爲特定之特 性，並藉由此來構築一種適合於裝置特性差、周圍噪音準 ' 位等的諸條件之雜音消除系統。 又，本發明，係以依據由參考麥克風所取得之資訊和 ❹ 從誤差麥克風所取得之資訊來對雜音消除處理部之特性作 控制一事爲其特徵。亦即是，係以藉由參考麥克風或是誤 差麥克風所觀察到之訊號爲依據，來在雜音消除控制部中 決定因應於周圍狀況之特性參數，並將此賦予至雜音消除 處理部者。 特別是，在本發明中，其特徵爲：藉由於雜音消除處 理部處，採用可藉由數位訊號來作控制之類比元件，而能 夠在保持類比元件之回應特性的同時，亦在控制參數之決 © 定中而導入具備有安定性與精密性之數位控制。於此情況 ’作爲雜音消除後之音樂訊號的輸出手段，係使用如同耳 _ 機、頭戴式耳機又或是耳機揚聲器一般之與耳朵間的距離 爲短者，藉由此，而有效地利用類比元件之迅速的回應性 〇 例如，作爲雜音消除處理部之構成機器，係採用電子 衰減器來構築噪音控制訊號之增益或是帶域限制濾波器。 進而，在本發明中，其特徵爲：當對於1個的雜音消 g系統而可作選擇地連接有特性爲相異之複數的頭戴式耳 -10- 200926138 機或是耳機的情況時，藉由從外部來在雜音消除處理部中 設定適合於於各頭戴式耳機或是耳機之參數，能夠維持原 本之電路構成，而進行適合於各裝置之特性的雜音消除。 〔發明之效果〕 在本發明中，係成爲能夠以參考麥克風之資訊爲依據 ，而進行適應於周圍之噪音準位的噪音控制，在高準位之 〇 噪音時，係使噪音壓抑量上升’而當低噪音準位時，則噪 音壓抑量係亦下降，藉由此，能夠對壓迫感或是不必要之 雜音作抑制。又，特別是在耳機所致之主動雜音消除系統 中’藉由亦對誤差麥克風之資訊作考慮，能夠構築出對個 人差或是裝著狀態之相異有所因應的主動式雜音消除系統 〇 若藉由本發明，則藉由於前授處理部或是後授處理部 處，採用可藉由數位訊號來作控制之類比元件，而能夠實 ® 現一種在保持類比元件之回應特性的同時，亦在訊號控制 部之特性決定中而導入具備有安定性與精密性之數位控制 的系統。 又，藉由從外部來在控制濾波器處設定適合於各別之 雜音消除耳機的參數，在至今爲止均係藉由對於各裝置而 分別進行了專用之調整（tuning )的電路來構成之雜音消 除頭戴式耳機系統中，係成爲能夠在維持相同之系統的同 時，而對各種之裝置作處理。 -11 - 200926138 【實施方式】 (1 )用語之說明 針對在本實施形態中所使用之用語作說明。 (a )參考麥克風（噪音源資訊取得感測器） 係爲設置在耳機之外側的麥克風，由於原則上係並不 包含耳機輸出音，因此，可以考慮係對到達至耳朵之噪音 本身作拾取者。此係爲在前授處理部處作爲取得噪音源之 0 資訊的感測器而起作用者。 (b )誤差麥克風（控制狀態評價感測器） 係爲設置在耳機之內側（耳朵側）的麥克風，並作爲 亦包含有「耳栓效果」之實際上所聽到的聲音而作處理。 由於亦包含有從耳機而來之音樂資訊，因此，若是以誤差 麥克風之收錄音爲依據而進行雜音消除處理，則其結果， 音樂訊號係亦會被消除。此係爲在後授處理部處作爲控制 狀態評價感測器而起作用者。 〇 ( C )頻率回應控制部與增益控制部 兩者均係爲雜音消除處理部3、13之構成構件，並爲 可進行數位控制之類比元件。增益控制部，係爲藉由電子 音量器、電子衰減器等所構成。頻率回應控制部，雖係進 行相位之統籌處理，但是，基本上係爲藉由電容器C與電 阻R所構成之1次乃至2次程度的濾波器。藉由使用梯形 電阻或是開關來對此電阻R之値作切換，而將作爲濾波器 之截頻或是衰減量設爲可變。此頻率回應控制部，係爲僅 使外部雜音之控制帶域內的成分通過（將控制頻率帶域外 -12- 200926138 之頻率成分遮斷）、而使不包含外部雜音或者是控制帶域 外之頻率帶域成爲不通過者。 (d)前授處理部 係爲本發明之雜音消除處理部的其中一例，將藉由參 考麥克風所輸入之噪音參考訊號，通過對直到耳機輸出部 爲止之傳導函數作了預想•近似之濾波器，而以逆相位來 輸出，藉由此，來對噪音作壓抑。實際上，由於相較於由 0 參考麥克風處之檢測起直到輸出爲止，音波係會先到達輸 出部，因此，作爲處理，係僅進行到對低頻域作壓抑。 (e )後授處理部 係爲本發明之雜音消除處理部另外一例，將藉由誤差 麥克風所輸入之訊號設爲逆相位而作輸出，並藉由此來對 噪音作壓抑。此時，由於係會產生由於電性音響變化系所 致之延遲，並引起嘷鳴（howling)，因此，一般而言， 係以不會使此些之延遲造成問題的程度，而限定在低頻來 G 作輸出。 (f) 類比/數位（A/ D )變換、數位/類比（D/ A )變換 如同前述一般，在前授、後授之兩者的處理中，均會 由於延遲而使控制可能帶域受到侷限。處理系之延遲特性 ，係與雜音消除性能有極深的相關，由於A/D變換或是 D/A變換所伴隨之延遲，現狀而言，處理系係藉由類比 電路而被構成。 (g) 音樂訊號 -13- 200926138 係爲從音源所輸出之各種訊號的總稱，而包含有聲音 、音樂、效果音等。亦稱爲原音。此係並不被限定於被作 錄音並作再生者。 (h )輸出手段 作爲在本發明中所使用之輸出手段，係將如同耳機、 頭戴式耳機或是耳機揚聲器一般之輸出手段與耳朵間之距 離爲短者爲對象。於此情況，所謂耳機揚聲器，係指像是 〇 聆聽椅（在兩耳之後方近旁安裝有揚聲器之椅子或是車的 座位）等，在耳朵之極近旁處作聆聽之揚聲器。 (2 )主觀評價所致之控制參數的決定 關於在本發明中之考慮有主觀評價的控制參數之決定 ，係藉由主觀評價，來對於所想定之噪音模式訊號而決定 在各種噪音準位時之最能感到舒適感之控制參數。於主觀 評價中，係包含有以下一般之要素。 〇 ( a )雜音控制所致之壓抑效果的主觀性之評價量 (b)在控制中所需要之電路上的寄生雜訊、或是與 在控制時所想定之傳導函數間的誤差、又或是在後授迴圈 中對增益上升之帶域等，由於進行控制所新產生之噪音。 (c )由於控制所感到之壓迫感或是逆相位感。 在此些之要素亦成爲包含對象之各系統中，決定在所 期望之噪音準位下最能夠感到舒適性之控制參數。雖然（ a ) 、（ b )係可作爲物理量來作測定，但是，關於（c ) ，在物理量與心理量之明確的關係性上，係並未成爲完全 -14 - 200926138 明瞭。又，（a)與（b)，由於多有著若是使其中—方減 少則另外一方係會增加之交易（trade off )性的關係，又 或是係爲非常複雜之系統，因此，若是以人類之舒適性作 爲評價基準而看待，則與（c)同樣的，要算出最適當之 平衡一事係爲困難。在此種背景下’於本發明中’係導入 主觀評價。 以此些之主觀評價爲依據，而進行下述一般之控制。 0 ( 1 )以所得到之各參數爲依據，而作成控制表，並 配合於雜音準位而設定適當之値。 (2 )以所得到之各參數爲依據，而計算出近似直線 、近似曲線，並逐次計算出對應於雜音準位之値而作設定 〇 (3)將所得到之各參數或是（考慮有個人差異而使 其具備有變化幅度）接近之値作爲控制表而作成，並讓利 用者從該些中來作選擇並利用。 ❹ 於圖10之曲線圖中，展示主觀評價曲線之其中一例 與近似曲線例。在此曲線圖中，後述處理部之控制增益（ 黑圏記號）與前授處理部之控制增益（黑方塊記號），係 對於各環境雜音準位的每一個，而將控制增益準位作增減 ’並將聽取者判斷爲最適當的値，決定爲該環境雜音準位 之控制增益準位。而後，根據在各環境準位之每一者中所 被決定之控制增益準位，來計算出決定控制參數之近似曲 線。 又’於圖1 1中，展示採用有此種主觀評價之控制系 -15- 200926138 統例的流程。在此圖1 1之控制系統中，首先，作爲準備 階段’係對聽取者所裝著之雜音消除耳機作辨識（步驟1 )’而後，藉由控制表’而進行可控制之帶域寬幅、傳導 特性的讀出.設定（步驟2)。 接下來，使裝著有該雜音消除耳機之聽取者，在雜音 環境中’藉由前授處理部之參考麥克風或是後授處理部之 誤差麥克風來對身爲控制對象之雜音訊號的頻率分佈作觀 Q 察（步驟3 )，並檢測出在藉由前述控制表而可進行控制 之頻率帶域內，是否有能量之分佈不均（是否僅包含有多 數之某一特定頻率帶域下的雜訊成分）（步驟4)。 在存在有能量之分佈不均的情況時，係以使在後授處 理部或是前授處理部中之控制濾波器的頻率特性近似於雜 音訊號之特性的方式來作更新（步驟5)，而當並不存在 有能量分佈不均的情況時，則各處理部之控制濾波器係被 設定爲可進行控制之範圍的全區域（步驟6 )。而後，配 〇 合於雜音訊號之控制帶域的能量之量（環境雜音準位）， 來對利用圖1 〇之主觀評價曲線圖等所製作之心理評價値 的表作參考，並對各處理部之控制濾波器的增益作更新（ 步驟7 )。 之後，伴隨著雜音消除耳機之使用的繼續，而恆常地 繼續進行對控制對象（雜音訊號）之頻率分佈的觀察’並 反覆進行前述步驟3〜步驟7’藉由此’而進行以配合有 雜音訊號之頻率分佈不均與控制帶域之能量之量的主觀評 價爲依據之控制濾波器的增益之更新’而實行了考慮有主 -16- 200926138 觀評價之雜音消除處理。 於此情況，當將使用有參考麥克風1之前授處理與使 用有誤差麥克風11之後授處理作倂用的情況時，在前述 圖11中之步驟7之後，係被實行有下述一般之步驟8、9 〇 亦即是，一般而言，係將耳栓之類的沒有主動作用之 消音裝置稱爲被動雜音消除系統，而將使用有揚聲器等之 〇 進行積極的控制之系統稱爲主動雜音消除系統，但是，當 由於頭戴式耳機或是耳機等所致之被動雜音控制而造成在 參考麥克風與誤差麥克風處之音場中產生有大的差距的情 況時，藉由誤差麥克風11來對雜音特性作觀察並對參數 設定作更新之手法’係可能會具備有更高之實用性。或者 是’當僅藉由誤差麥克風11來進行雜音消除時，係有必 要以誤差麥克風訊號爲依據來對參數設定作更新。 又’在藉由參考麥克風1以及誤差麥克風n之倂用 ^ 來對個人差（於此情況中之個人差，係不僅是各人之在外 耳道處的傳導函數之差異，而亦廣義地包含有該情形下之 耳機的裝著方法等之與被作一般化參數設定之資訊間的實 際的差異）作修正的手法中，係成爲能夠使用噪音訊號或 是音樂訊號’來將從參考麥克風i起直到誤差麥克風U 之傳導函數、從輸出手段10起直到誤差麥克風11之傳導 函數藉由同定濾波器16來作同定（步驟8)，並在前授 處理部3中’藉由前者來對其之控制濾波器之頻率特性以 及增益特性作更新（（微）調整）而在後授處理部13中 -17- 200926138 ，藉由後者來對其之控制濾波器之頻率特性以及增益特性 作更新。 (3 )第1實施形態 ' 根據圖5，對考慮有前述一般之主觀評價的本發明之 雜音消除系統的第1實施形態作說明。此第1實施形態， 係爲僅使用被設置在前授處理部處之參考麥克風而進行雜 φ 音消除者。 於圖1中，符號1係爲參考麥克風，並將外部雜音作 爲類比訊號而輸入。此參考麥克風之輸出，係經由麥克風 擴大器2，而被輸出至前授處理部3與雜音消除控制部4 中。 前授處理部3，係具備有頻率回應控制部3a與增益 控制部3 b。在本實施形態中，此些之頻率回應控制部3 a 與增益控制部3b，係由可作數位控制之類比元件所構成 〇 ，並根據從雜音消除控制部4而來之數位資料，來將外部 雜音之特定的頻率帶域作截除，或是作衰減。 從麥克風擴大器2而來之外部雜音，係經由A/D變 換器5而被作數位資料化，而後，被輸入至雜音消除控制 部4中。此外部雜音檢測部4，係具備有：用以對所輸入 之外部雜音的大小與何者之頻率帶域的能量係爲大一事作 觀察之外部雜音檢測部4a、和根據此外部雜音檢測部4a 之檢測結果而從控制表6來取得特定之參數的參數取得部 4b、和根據此參數取得部4b所取得之參數，來將控制資 -18 - 200926138 料輸出至前述頻率回應控制部3a與增益控制部3b中之資 料輸出部4c。 在前述前授處理部3之輸出側，係被設置有加算器7 ’在此加算器7中，係將從前授處理部3所輸出之雜音消 除訊號和從音源8而來之音樂訊號作加算。於此情況，在 本實施形態中，作爲音樂訊號，由於係使用數位訊號，因 此，從音源8而來之音樂訊號，係在藉由D/ A變換器9 〇 而被變換爲類比資料之後，再在加算器7中與同樣身爲類 比資料之雜音消除訊號作加算，並從揚聲器、頭戴式耳機 或是耳機等之輸出手段10而輸出。 另外，在本實施形態中，亦可代替經由參數取得部 4b來從控制表6取得參數，而設置參數演算部，並對應 於以檢測部4 a所檢測出之外部雜音的準位或是能量分佈 ，而於每次實行中對參數作演算。又，亦可列舉出依據控 制表6所得之近似曲線的使用、或是依據控制表6來從噪 〇 音環境狀況來推定出最適當之參數的手法。 在具備有此種構成之第1實施形態中，藉由參考麥克 風1所收音之外部雜音，係被輸入至前授處理部3與雜音 消除控制部4中。在雜音消除控制部4中，係在其之外部 雜音檢測部4a處，對被輸入的雜音準位以及該些之在頻 率軸上的能量分佈作觀察，並在參數取得部4b處，計算 出適合於所輸入之外部雜音的參數，或是從控制表6而取 得預先所計算出之參數’而將此從資料輸出部4c來輸出 至前授處理部3處。 -19- 200926138 在前述處理部3中，係根據從前述雜音消除控制部4 所取得之參數’而對頻率回應控制部3 a以及增益控制部 3b作控制，並將從參考麥克風1而經由麥克風擴大器2 所輸入之外部雜音’通過對直到輸出手段1〇爲止之傳導 ' 函數作了預想•近似後之濾波器，而以逆相位作輸出’藉 ' 由此，來施加雜音消除處理。 於此情況，在本發明中，構成頻率回應控制部3a以 Q 及增益控制部3b之機器，雖然係爲將作爲類比訊號而被 輸入之外部雜音直接以類比資料之狀態而作處理之類比元 件，但是，該控制量，係爲經由從雜音消除控制部4而來 之數位訊號而被決定者。故而，若藉由本實施形態，則在 頻率回應控制部3a以及增益控制部3b處之對於外部雜音 的消除處理其本身，係並不需要進行A/D變換或是D/ A變換，因此，具有能夠使雜音消除處理之延遲成爲最小 限度之優點。 (4 )第2實施形態 根據圖6，來對本發明之第2實施形態作說明。此第 2實施形態，係爲將參考麥克風與誤差麥克風作了倂用者 。另外，針對與前述第1實施形態相同之構成，係省略其 說明。 於圖6中，11係爲誤差麥克風，此誤差麥克風n， 係爲對聽取者所實際聽到之聲音（被作了雜音消除的聲音 )作採取者’而被設置在頭戴式耳機、耳機等之輸出手段 -20- 200926138 ίο的近旁。 從此誤差麥克風11而來之誤差麥克風訊號，係經由 麥克風擴大器12而被放大，而後，被輸出至後授處理部 13與前述雜音消除控制部4中。於此情況，誤差麥克風 ' 訊號，對於後授處理部1 3，係作爲誤差麥克風1 1所輸入 之類比訊號來被輸入，而對於前述雜音消除控制部4，係 經由A/D變換器14，而作爲數位資料來被輸入。 〇 前述後授處理部13，係與前述前授處理部3同樣的 ，具備有頻率回應控制部13a與增益控制部13b，此些之 控制部1 3a、1 3b，係爲藉由從雜音消除控制部4而來之 數位資料而被作控制之類比元件，此點，係和前述前授處 理部3爲相同。 在此第2實施形態中，雜音消除控制部4，係除了前 述第1實施形態之功能以外，亦根據誤差麥克風訊號，而 產生用以對後授處理部1 3之各控制部1 3 a、1 3 b作控制之 ® 數位訊號。亦即是，外部雜音檢測部4，係具備有：根據 外部雜音檢測部4a之檢測結果而從控制表6來取得特定 之參數的參數取得部4b、和根據此參數取得部4b所取得 之參數，來將控制資料輸出至前述頻率回應控制部3a與 增益控制部3b中之資料輸出部4c，並藉由從此資料輸出 部4c而來之訊號，來對後授處理部1 3之頻率回應控制部 1 3 a與增益控制部1 3 b作控制。 然而’在如同第2實施形態一般之後授型的雜音控制 系統中，係在音源與其之D/A變換器9之間，配置修正 -21 - 200926138 濾波器15。亦即是，在藉由誤差麥克風11所收錄之誤差 麥克風訊號中，由於係存在有被消除了的雜音之殘留份與 從音源8而來之音樂訊號，因此，在後授處理部13中， 若是以此誤差麥克風訊號爲依據而產生逆相位之雜音消除 ' 訊號，則從音源而來之音樂訊號亦會成爲被消除。 因此，此修正濾波器15，係從雜音消除控制部4，而 取得相當於在後授處理部13處之音樂訊號的消除份之資 Q 料，並因應於其，而對於從音源8而來之音樂訊號施加濾 波處理，藉由此，就算是當從後授處理部13而被施加有 相當於音樂訊號之雜音消除訊號的情況時，亦能夠以不損 及原本之音樂訊號的方式，而對從音源8而來之音樂訊號 作修正。 在此種於第2實施形態中所採用之後授型的雜音控制 系統中，係有必要在每一次之參數的更新中，對修正濾波 器15作更新。因此，在第2實施形態中，係在雜音消除 φ 控制部4中，設置同定濾波器16與逆相位產生器17，並 對從輸出音源8而來之音樂訊號（作爲其中一例，係從修 正濾波器15之後段來取得）與從誤差麥克風11而來之誤 差麥克風訊號作比較，而將對於雜音消除系統之音源的回 應，使用同定濾波器16(例如’LMS等之適應濾波器） 來作同定，在根據此同定後之資訊，來在逆濾波器產生器 1 7中，亦將該逆濾波器作近似。藉由使用如此這般所得 到之近似逆濾波器來將修正濾波器15作更新，而成爲能 夠在聽覺上如同聽到原本之音源訊號一般地來進行前處理 -22- 200926138 如同上述一般，在第2實施形態中，從誤差麥克風 11而被輸入至雜音消除控制部4中之訊號，由於係作爲 耳機之人體側、亦即是實際上在耳中所聽到之聲音而作處 理’因此’作爲欲將雜音消除之機構，係會被觀察到所謂 的「消除遺漏」。另一方面，若是注目於從音源8而來之 音樂訊號’則係能夠透過誤差麥克風1 1，而觀察到「（ 〇 經由雜音消除處理而變質後之）實際上在耳中所聽到之音 樂」。又’在雜音消除控制部4中，由於係從修正濾波器 15之輸出側而亦被輸入有再生音之音樂訊號，因此，只 要對誤差麥克風Η之聲音作觀察，則亦能夠判明音樂係 被作了如何之變質。 而’作爲雜音消除控制部4之動作，係從原音（音樂 訊號）與誤差麥克風訊號，來藉由LMS等之同定濾波器 16而求取出訊號之變化量（從揚聲器起朝向耳中之系統 〇 的脈衝回應）’並根據從此同定濾波器16而來之輸出， 來在逆相位產生器17中，產生其之逆濾波器（將通過了 前述之系統時的傳導函數設爲“ 1 ”，亦即是，正好將前述 之系統作消除一般之濾波器）的脈衝回應資訊，並藉由此 來將修正濾波器15作更新。其結果，若藉由第2實施形 態，則係具備有可使音質修正濾波器1 5追隨於雜音控制 參數之變化而作變更的優點。 另外，在第2實施形態中，當數位控制類比部之參數 係爲有限之値的情況時，則亦能夠以對於雜音消除部之參 -23- 200926138 數作爲依據，來參考控制表6而對修正濾波器15作更新 〇 進而，在此第2實施形態中，係藉由將以誤差麥克風 11所檢測出之誤差麥克風訊號中的外部雜音成分輸入至 前述雜音消除控制部4之外部雜音檢測部4a中，而對於 在後授處理部1 3中之控制濾波器的參數作控制。於此情 況，如同前述一般之從誤差麥克風11而藉由麥克風放大 〇 器12來作放大，而後再藉由A/D變換器14而被作了數 位訊號化之資訊，係爲「在耳朵中所直接聽到的應作控制 之雜音.+到達耳朵之音樂訊號」。 但是，在外部雜音資訊之檢測中，係要求有能僅將音 樂訊號成分作取出。因此，在此第2實施形態中，係在從 數位音源8而來之音樂訊號中，將在音質修正濾波器之項 中所計算出的從輸出手段10起直到誤差麥克風11的傳導 函數作了同定後之濾波器18作卷積（convolution)，並 〇 在加算器19中，從經由A/D變換器14而被變換爲數位 訊號後的誤差麥克風訊號來作減算。其結果，係成爲僅有 誤差麥克風訊號中之外部雜音成分被輸入至外部雜音檢測 部4a中，而能夠與從參考麥克風1而來之外部雜音同樣 的，對構成後授處理部1 3之頻率回應控制部1 3 a以及增 益控制部13b之參數作決定。 (5 )第3實施形態 圖7，係爲展示本發明之第3實施形態者，代替前述 -24- 200926138 第1實施形態之根據經由參考麥克風所取得的外部雜音來 決定參數，係因應於裝置之特性或是個人之耳朵的特性、 並更進而因應於雜音之發生環境，而從預先所準備並設置 之複數的參數中，選擇使用者所希望之參數。 亦即是，在第3實施形態中，係在雜音消除控制部4 中，代替外部雜音檢測部4a，而被設置有參數選擇開關 21，並被構成爲能夠經由此參數選擇開關21，來對參數 0 取得部4b之從控制表6所取得的參數作選擇。 因此，在第3實施形態中，係藉由因應於所使用之頭 戴式耳機或是耳機等之裝置的形態或是特性而對開關21 作操作、或是配合於使用者個人之耳朵的特性、嗜好、或 是噪音之狀況（雜音之特性）而對開關21作操作，來選 擇所期望之參數，並對前授處理部3之頻率回應控制部 3a與增益控制部3b作控制，而能夠進行適當之雜音消除 處理。 ❹ (6 )第4實施形態 圖8，係爲展示本發明之第4實施形態者，此第4實 施形態，係爲將在前述第3實施形態中所展示之參數選擇 開關2 1所致的前授處理，與在第2實施形態中所示之後 授處理作了組合者。 經由此種構成之第4實施形態，與前述第2實施形態 同樣的，成爲能夠在將從誤差麥克風11而來之訊號作後 授的同時，進行配合於裝置或是周圍之噪音狀況的參數之 -25- 200926138 決定。 (7 )第5實施形態 圖9，係爲展示本發明之第5實施形態者，此第5實 施形態，係並不使用參考麥克風1，而僅經由誤差麥克風 11來進行外部雜音之檢測，同時，進·行修正濾波器1 5所 致之音質的修正者。 〇 另外，在圖9中，雖然與第1實施形態同樣的’係根 據經由誤差麥克風11所檢測出之外部雜音來決定參數， 但是，係亦可如同前述第3實施形態或第4實施形態一般 ，經由參數選擇開關21來決定參數。 在此種構成之第5實施形態中，亦與前述各實施形態 同樣的，成爲能夠藉由將從誤差麥克風11而來之訊號作 後授，來進行配合於周圍之噪音狀況的參數之決定。 〇 ( 8 )各實施形態所致之效果 具備有前述一般之構成以及作用的各實施形態之效果 ，係如下述所示一般。 在前述之實施形態中，在被要求無延遲之回應性的雜 音消除部中，係僅使用（數位控制）類比訊號處理，而在 延遲並不會成爲問題但是卻需要正確之修正濾波器的音源 再生部中，係僅施加有數位訊號處理系。其結果，在音源 再生部中，係成爲能夠將包含有身爲在類比濾波器中原理 上係爲不可能構成的非因果性之要素的瀘波器作組入，而 -26- 200926138 藉由與在頭戴式耳機中之立體再生系統作倂用，而成爲能 夠在進行雜音消除的同時而建構雙耳系統。 進而’藉由將修正濾波部以數位來構成並與適應濾波 器作組合’而實現了在爲了維持虛擬系統而需要進行複雜 之處理的雜音消除部之係數更新中的再生音用之修正濾波 器的更新。 前述之實施形態，係以在被廣泛作採用之類比構成的 〇 雜音消除頭戴式耳機系統中，將數位控制作導入，並藉由 將系統之特性•增益從外部而設爲可變，來在同一之系統 中產生適合於裝置狀況或是噪音狀況之雜音消除訊號一事 爲特徵。特別是，藉由採用可進行數位控制之類比元件， 而能夠實現一種在具備有類比之回應性的同時亦可進行數 位控制之系統。 在前述實施形態中，由於增益控制係爲1區塊1參數 ，而頻率回應控制亦係藉由類比元件所構成而一般係成爲 〇 1乃至2左右之數個參數，因此，亦能夠使用數分支（tap )之LMS或是微擾法（Perturbation methods)等之適應 處理來進行最適化。故而，能夠建構一種：相較於具備有 被廣泛進行有硏究之全數位化適應訊號處理的雜音消除， 所需之計算量係遠爲少，且具備有在數位中所不存在之回 應特性，而可供以實用的系統。 【圖式簡單說明】 〔圖1〕對本發明之雜音消除系統的第1特徵作說明 -27- 200926138 之區塊圖。 〔圖2〕對本發明之雜音消除系統的第2特徵作說明 之區塊圖。 〔圖3〕對本發明之雜音消除系統的第3特徵作說明 之區塊圖。 〔圖4〕對本發明之雜音消除系統的第4特徵作說明 之區塊圖。 〇 〔圖5〕展示本發明之雜音消除系統的第1實施形態 之區塊圖。 〔圖6〕展示本發明之雜音消除系統的第2實施形態 之區瑰圖。 〔圖7〕展示本發明之雜音消除系統的第3實施形態 之區塊圖。 〔圖8〕展示本發明之雜音消除系統的第4實施形態 之區塊圖。 ❹ 〔圖9〕展示本發明之雜音消除系統的第5實施形態 之區塊圖。 〔圖10〕展示在本發明中之主觀評價曲線（plot)之 其中一例與近似曲線例的圖表。 〔圖11〕展示在本發明中之控制系統的其中一例之 流程圖。 【主要元件符號說明】 1 :參考麥克風 -28- 200926138 2 :麥克風擴大器 3 :前授處理部 3 a :頻率回應控制部 3 b :增益控制部 4 :控制部 ' 4a :外部雜音檢測部 4b :參數取得部 Q 4c :資料輸出部 5 : A/D變換器 6 :控制表 7 :加算器 8 :音源 9 : D/A變換器 I 〇 :輸出手段 II :誤差麥克風 〇 12:麥克風擴大器 1 3 :後授處理部 1 3 a :頻率回應部 1 3 b :增益控制部 14 : A/ D變換器 Η :音質修正濾波器 1 6 :同定濾波器 1 7 :逆濾波器產生器 18 : D/Α變換器 -29 200926138 1 9 :加算器 2 1 :參數選擇開關200926138 Nine, the invention belongs to the technical field of the invention. The present invention relates to a listening machine suitable for use in a speaker such as a noise canceling headphone or a headphone, a headphone speaker, etc., which is generally placed near the ear. Noise control system. [Prior Art] 〇 In recent years, with the popularity of portable music players, various noise-eliminating headphones (headphones) have become popular. The noise canceling headphone is located near the listening point, and is provided with a noise source information detecting sensor or a control state evaluation sensor. If it is the former case, by using the sensor from the sensor until the listening point The transmission characteristics are simulated filters to attenuate the noise level of the listening point; in the latter case, a regression filter that does not cause howling occurs, and by adding a reverse phase to the sound field The signal is used to attenuate the noise level of the listening point. © Most of the noise-removing headsets currently on the market are analog-like noise-eliminating headsets. In this way, by the microphone that is incorporated into the headset, the surrounding sound (relative to the reproduced sound, because this is a noise, therefore, referred to as "noise" below), is captured. The phase of the captured noise is reversed and added to the reproduced signal from the player. In this configuration, the noise that intrudes into the inside of the headset from the outside is cancelled by the signal whose phase is reversed, and only the reproduced signal from the player enters the user's ear. in. -4- 200926138 Recently, 'although there are also proposed digital noise cancellation headphones', but 'generally' noise-removing headphones are a configuration that closely connects the sensor to the listening point of the speaker. However, many delays caused by converters exceed the arrival time difference of the sound waves, and large delay times are required in A/D conversion and D/A conversion. Therefore, the current situation is digital signal processing. Almost unacceptable. In the prior art of the analog-type noise canceling headphone, for example, the invention described in Patent Document 1 or Patent Document 2 exists. Further, as a prior art of the digital noise canceling headphone, for example, the invention described in Patent Document 3 exists. [Patent Document 1] Japanese Laid-Open Patent Publication No. Hei 11-237889 (Patent Document No. Hei. Problem] Therefore, in the above-mentioned general prior art, there are the following general problems, and it is expected that such problems can be solved. (1) Stereoscopic reproduction under noise In the existing binaural system for the purpose of three-dimensional reproduction, there is a certain degree of noise, and the audio image cannot be positioned due to the masking effect. Problems. In today's widely used noise-eliminating headsets and noise-canceling earphones, the correction filter for source-5-200926138 is constructed by a low-order analog filter and stays only in the region of attenuation. Emphasis on the use of binaural systems that require detailed phase information is still difficult. (2) Attention to the comfort of the existing noise-eliminating headphones, noise-removing headphones, or related research reports, etc., almost all pay attention to the attenuation of the noise that is the object of control By. However, as a matter of reality, in the case of low noise, it is known that if excessive noise is removed, there is a feeling of oppression or discomfort, and sometimes, in the noise cancellation circuit itself, There is a noise that is parasitic in the microphone amplifier or the like. Therefore, if the noise cancellation amount is set to the limit regardless of the environmental noise level, the comfort is not necessarily improved. That is to say, in the active noise canceling system caused by the previous headphone or earphone, the amount of suppression of the noise is sometimes felt to have a pressure of €1 or a sense of reverse. Further, in most of the noise canceling headphones, it is found that the noise level outside the suppressed target band of the high frequency or the like rises due to the limitation of the number of times of filtering by the control filter unit. In particular, in the active noise cancellation system caused by the previous fixed filter, problems such as a feeling of pressure in a low-level noise state or an increase in the level of noise can be found. (3) Adaptation processing for non-periodic sounds caused by parameter limitation In the widely popular noise canceling headphones, noise canceling headphones -6 - 200926138 'The whole analogy system used, in composition, on the characteristics Updating is difficult. In addition, in the academic field or in the ANC (active noise can cal) of other structures such as ducts, the LMS (Least Mean Square) algorithm, which is widely used, has digital filtering. The noise cancellation system of the device is configured in a configuration in which the reference microphone is extremely closely connected to the speaker and the error microphone as in the case of the headset, and the delay is accompanied by the A/D conversion and the D/A conversion. There is a high sampling 'and 'a large increase in the sampling frequency, which will increase the amount of calculation.' In the non-periodic noise control, many systems cannot be put into practical use. (4) Adaptation to personal differences In the headphone-type noise cancellation system, basically, in order to obtain the best results in the two methods of pre-granting and post-granting, it is necessary to respond to the state of the listener. The processing is performed from the reference microphone until the transmission function of the speaker®, from the speaker to the transmission function of the error microphone. However, in the prior art, only the transfer function prepared by the system side is used and the processing corresponding thereto is performed, and therefore, the adaptation process for each listener cannot be performed. (5) Adaptation to the device difference In the existing noise canceling headphone/hybrid-eliminating earphone system, since it is composed of one processing system in one device, it is necessary to separately prepare various types. The device for eliminating noise of the characteristics, 200926138 It is impossible to connect and use various noise canceling devices, for example, in a portable music player of one model. The present invention has been made in order to solve the problems of the prior art as described above, and an object thereof is to provide a feature capable of changing a coefficient while maintaining a noise canceling function, and thereby In the update item, not only the attenuation of the object noise, but also the discomfort under low noise is incorporated into the evaluation function, and the noise control system that uses comfort as the evaluation reference is realized. Another object of the present invention is to provide a proposal for limiting an update item for an adaptive processing segment of an evaluation function to a plurality of segments, and constructing a noise cancellation system by using analog components that can be digitally controlled, thereby providing With high-speed responsiveness, it is also possible to update parameters with extremely realistic calculations, and to use a practical system with the advantages of analogy and digital. In particular, the present invention is one of the embodiments in which the above-described general purpose can be achieved for an earphone, a headphone, an earphone, or the like that is used in the vicinity of the ear. The reason is as follows. That is to say, the so-called active noise control for the sound field control of various noise fields and creating a sound environment that is comfortable for humans can be said to be affected by the development of digital signal processing represented by DSP. One of the technologies that make practical use of rapid progress. For example, from the noise cancellation at the duct of an air conditioner or the like, or the noise cancellation headset, until the noise control in the car or the noise reduction of the factory or the highway, regardless of the size, the study is tied to get on. However, in the earphones, the headphone -8 - 200926138 type earphones, and the noise control at the headphone speaker, in the special case, the following general features can be cited. (n from the scale of the system being constructed, 'generally' the distance between the reference microphone-outer ear canal and the reference microphone-speaker is extremely short, and therefore, due to the A/D conversion or the D/A conversion. With the delay, the configuration of the control filter caused by the digital signal processing has many disadvantages. © (2) Since the control target point is near the tympanic membrane, depending on the amount of control, there is a feeling of oppression. Or, it is a feature that is sensitive to new noise generated by the control. If a method that is widely tried and evaluated only by the error volume observed by the error microphone is used, the evaluation will be made. In the present invention, it is not equivalent to the comfort that humans actually feel. In the present invention, it corresponds to (1) and (2), and the proposal® (a) digital signal is used only in control. And the signal system only passes through the analog path, and the noise correction in the earphones and the like is specialized to the digital control and analog noise cancellation system which has the advantages of digital and analogy. (b) Not only the noise processing The control amount of the rate is used as an evaluation criterion, and is also adapted based on an evaluation function including the subjective evaluation of the comfort of the listener. [Means for Solving the Problem] In order to achieve the above object, the noise control of the present invention The system, which features -9-200926138, uses a variable component such as an electronic volume or an attenuator to eliminate noises that constitute a pre-processing unit or a post-processing unit. The filter (frequency response control unit) and the gain amplifier (gain control unit) of the processing unit are controlled to have specific characteristics from the outside, thereby constructing a condition suitable for device characteristic difference, ambient noise level, and the like. Further, the present invention is characterized by controlling the characteristics of the noise canceling processing unit based on the information obtained by the reference microphone and the information obtained from the error microphone. Based on the signal observed by the reference microphone or the error microphone, the noise cancellation control unit determines the condition corresponding to the surrounding conditions. The parameter is given to the noise canceling processing unit. In particular, in the present invention, it is characterized in that, by the noise canceling processing unit, an analog component that can be controlled by a digital signal can be used. While maintaining the response characteristics of the analog components, digital control with stability and precision is also introduced in the decision of the control parameters. In this case, the output signal of the music signal after the noise is eliminated is used as the ear. _ The machine, the headset, or the earphone speaker generally have a short distance from the ear, thereby effectively utilizing the rapid response of the analog component, for example, as a constituent machine of the noise cancellation processing unit, An electronic attenuator is used to construct a gain of a noise control signal or a band limiting filter. Further, in the present invention, it is characterized in that a characteristic is phase-connected for one noise cancellation system. Different types of headphones-10-200926138 In the case of a machine or an earphone, it is suitable for each head wear in the noise cancellation processing unit from the outside. The parameters of the earphone or the earphone can maintain the original circuit configuration and perform noise cancellation suitable for the characteristics of each device. [Effect of the Invention] In the present invention, noise control adapted to the surrounding noise level can be performed based on the information of the reference microphone, and the noise suppression amount is increased when the noise is high at the high level. When the noise level is low, the noise suppression amount is also lowered, whereby the pressure feeling or the unnecessary noise can be suppressed. In addition, especially in the active noise cancellation system caused by headphones, by considering the information of the error microphone, it is possible to construct an active noise cancellation system that responds to differences in personal or installed status. According to the present invention, by using a pre-processing unit or a post-processing unit, an analog component that can be controlled by a digital signal can be used to maintain the response characteristics of the analog component. A system with digital control with stability and precision is introduced in the determination of the characteristics of the signal control unit. Further, by setting parameters suitable for the respective noise canceling headphones at the control filter from the outside, it has been conventionally constructed by a circuit that performs dedicated tuning for each device. In the elimination of the headphone system, it is possible to process various devices while maintaining the same system. -11 - 200926138 [Embodiment] (1) Description of Terms The terms used in the present embodiment will be described. (a) The reference microphone (noise source information acquisition sensor) is a microphone that is disposed outside the earphone. Since the earphone output sound is not included in principle, it is considered that the pickup of the noise reaching the ear itself can be considered. . This is a function of a sensor that acquires 0 information of a noise source at the pre-processing unit. (b) The error microphone (control state evaluation sensor) is a microphone provided on the inner side (ear side) of the earphone, and is treated as a sound actually heard by the "ear plug effect". Since the music information from the earphones is also included, if the noise cancellation processing is performed based on the recording of the error microphone, the music signal system is also eliminated. This is a function of the control state evaluation sensor at the post-processing unit. 〇 (C) Both the frequency response control unit and the gain control unit are constituent members of the noise canceling processing units 3 and 13, and are analogous components capable of digital control. The gain control unit is constituted by an electronic volume meter, an electronic attenuator or the like. The frequency response control unit performs phase-integration processing, but is basically a filter that is formed by the capacitor C and the resistor R once or twice. This resistor R is switched by using a ladder resistor or a switch, and the cutoff or attenuation amount of the filter is made variable. The frequency response control unit is configured to pass only components in the control region of the external noise (interrupting the frequency components of the control frequency band -12-200926138) so as not to include external noise or frequency outside the control band. The band becomes a non-passer. (d) The pre-processing unit is an example of the noise cancellation processing unit of the present invention, and a filter that is expected to be approximated by a conduction function up to the headphone output portion by a noise reference signal input from a reference microphone And output in reverse phase, whereby the noise is suppressed. In fact, since the acoustic wave system first reaches the output portion as compared with the detection from the 0 reference microphone until the output, the processing is performed only to suppress the low frequency region. (e) The post-processing unit is another example of the noise canceling processing unit of the present invention, and outputs a signal input by the error microphone as a reverse phase, thereby suppressing noise. At this time, since the delay due to the electrical acoustic system is caused and the howling is caused, in general, it is limited to the low frequency so as not to cause such delays. Let G be the output. (f) Analog/digital (A/D) transform, digital/analog (D/A) transform. As in the foregoing, in the processing of both pre- and post-delivery, the control may be subject to delay due to delay. Limitations. The delay characteristics of the processing system are deeply related to the noise cancellation performance. Due to the delay accompanying the A/D conversion or the D/A conversion, the processing system is constructed by an analog circuit. (g) Music signal -13- 200926138 is a general term for various signals output from the sound source, including sound, music, sound effects, and so on. Also known as the original sound. This system is not limited to being recorded and reproduced. (h) Output means As the output means used in the present invention, the distance between the output means of the earphone, the headphone or the earphone speaker and the ear is short. In this case, the earphone speaker refers to a speaker that is like a 聆 listening chair (a chair with a speaker installed near the ears or a seat of the car), and a speaker that is heard near the ear. (2) Determination of control parameters due to subjective evaluation The decision on the control parameters for which subjective evaluation is considered in the present invention is determined by subjective evaluation for various noise levels for the desired noise mode signal. The most control parameter for comfort. In the subjective evaluation, the following general elements are included. 〇(a) The subjective evaluation of the suppression effect caused by the noise control (b) the parasitic noise on the circuit required for control, or the error between the conduction function and the conduction function that is intended to be controlled, or It is the noise generated by the control in the band of the gain in the post-return circle. (c) The sense of oppression or reverse phase felt by the control. Among these systems, the elements that are included are the control parameters that determine the most comfortable comfort at the desired noise level. Although (a) and (b) can be measured as physical quantities, with regard to (c), the explicit relationship between physical quantity and psychological quantity has not become completely -14 - 200926138. Moreover, (a) and (b) have a trade-off relationship in which the other party is reduced if it is reduced, or a very complicated system, and therefore, if it is a human The comfort is regarded as the evaluation criterion, and it is difficult to calculate the most appropriate balance as in (c). In this context, 'in the present invention' is introduced into a subjective evaluation. Based on these subjective evaluations, the following general controls are performed. 0 (1) Based on the obtained parameters, a control table is created, and an appropriate level is set in accordance with the noise level. (2) Calculate the approximate straight line and the approximate curve based on the obtained parameters, and calculate the corresponding parameters corresponding to the noise level and set the parameters (3) or (considering Individual differences are made with a change range as a control table, and the user is selected and utilized from among them.于 In the graph of Fig. 10, one example of the subjective evaluation curve and an example of the approximate curve are shown. In this graph, the control gain (black mark) of the processing unit described later and the control gain (black square symbol) of the pre-processing unit are used to increase the control gain level for each of the environmental noise levels. Subtracting 'and determining the listener as the most appropriate 値 determines the control gain level for the ambient murmur level. Then, an approximate curve for determining the control parameter is calculated based on the control gain level determined in each of the environmental levels. Further, in Fig. 11, a flow using the control system -15-200926138 having such subjective evaluation is shown. In the control system of FIG. 11, first, as a preparation stage, the noise canceling earphones installed by the listener are identified (step 1)', and then the controllable band width is controlled by the control table. Readout of conduction characteristics. Set (step 2). Next, the listener who has the noise canceling earphone is used to detect the frequency distribution of the noise signal as the control object by the reference microphone of the pre-processing unit or the error microphone of the post-processing unit in the noise environment. Observing Q (step 3), and detecting whether there is uneven distribution of energy in the frequency band that can be controlled by the aforementioned control table (whether or not only a certain frequency band is included Noise component) (Step 4). When there is uneven distribution of energy, the frequency characteristic of the control filter in the post-processing unit or the pre-processing unit is updated to approximate the characteristics of the noise signal (step 5). On the other hand, when there is no unevenness in energy distribution, the control filter of each processing unit is set to the entire area in which the control can be performed (step 6). Then, the amount of energy (environmental noise level) in the control band of the noise signal is used to refer to the table of psychological evaluations made by using the subjective evaluation curve of FIG. 1 and the processing is performed. The gain of the control filter is updated (step 7). After that, with the continuation of the use of the noise canceling earphone, the observation of the frequency distribution of the control object (noise signal) is continuously continued, and the above steps 3 to 7 are repeatedly performed by the above. A noise cancellation process that considers the evaluation of the main-16-200926138 is performed based on the fact that the frequency distribution of the noise signal and the subjective evaluation of the amount of energy in the control band are based on the update of the gain of the control filter. In this case, when the reference microphone 1 is used before the processing and the error microphone 11 is used for the processing, after the step 7 in Fig. 11 described above, the following general step 8 is carried out. In other words, in general, a muffler that does not have an active action such as an ear plug is called a passive noise canceling system, and a system that uses a speaker or the like for active control is called active noise canceling. System, however, when there is a large gap between the reference microphone and the error microphone due to passive noise control caused by headphones or headphones, the error microphone 11 is used to correct the noise. The feature of observing and updating the parameter settings may have higher practicability. Alternatively, when the noise cancellation is performed only by the error microphone 11, it is necessary to update the parameter setting based on the error microphone signal. In addition, by using the reference microphone 1 and the error microphone n, the personal difference (in this case, the personal difference is not only the difference in the conduction function of each person at the external auditory canal, but also broadly includes In this case, the method of correcting the difference between the mounting method of the earphone and the information that is set as the generalized parameter is to use the noise signal or the music signal to start from the reference microphone i. Until the conduction function of the error microphone U, from the output means 10 until the conduction function of the error microphone 11 is determined by the same filter 16 (step 8), and in the pre-processing unit 3 'by the former The frequency characteristic and the gain characteristic of the control filter are updated ((micro) adjustment) in the post-processing unit 13-17-200926138, by which the frequency characteristics and gain characteristics of the control filter are updated. (3) First Embodiment A first embodiment of the noise canceling system of the present invention in consideration of the above-described general subjective evaluation will be described with reference to Fig. 5 . In the first embodiment, the noise canceling is performed using only the reference microphone provided in the pre-processing unit. In Fig. 1, symbol 1 is a reference microphone, and an external noise is input as an analog signal. The output of this reference microphone is output to the pre-processing unit 3 and the noise canceling control unit 4 via the microphone amplifier 2. The pre-processing unit 3 includes a frequency response control unit 3a and a gain control unit 3b. In the present embodiment, the frequency response control unit 3a and the gain control unit 3b are constituted by analog elements that can be digitally controlled, and based on the digital data from the noise cancellation control unit 4. The specific frequency band of the external noise is cut off or attenuated. The external noise from the microphone amplifier 2 is digitized by the A/D converter 5, and then input to the noise canceling control unit 4. The external noise detecting unit 4 is provided with an external noise detecting unit 4a for observing the magnitude of the external noise input and the frequency band of the frequency band, and the external noise detecting unit 4a. The parameter acquisition unit 4b that acquires the specific parameter from the control table 6 and the parameter acquired by the parameter acquisition unit 4b output the control element -18 - 200926138 to the frequency response control unit 3a and the gain. The data output unit 4c in the control unit 3b. On the output side of the pre-processing unit 3, an adder 7' is provided in the adder 7 to add the noise canceling signal output from the pre-processing unit 3 and the music signal from the sound source 8. . In this case, in the present embodiment, since the digital signal is used as the music signal, the music signal from the sound source 8 is converted into analog data by the D/A converter 9 ,. Then, in the adder 7, the noise canceling signal which is also analogous data is added, and is outputted from the output means 10 such as a speaker, a headphone or a headphone. Further, in the present embodiment, instead of acquiring parameters from the control table 6 via the parameter acquisition unit 4b, the parameter calculation unit may be provided to correspond to the level or energy of the external noise detected by the detection unit 4a. Distribution, and the parameters are calculated in each implementation. Further, the use of the approximate curve obtained in accordance with the control table 6 or the method of estimating the most appropriate parameter from the noise environment according to the control table 6 may be cited. In the first embodiment having such a configuration, the external noise received by the reference microphone 1 is input to the pre-processing unit 3 and the noise canceling control unit 4. The noise canceling control unit 4 observes the input noise level and the energy distribution on the frequency axis at the external noise detecting unit 4a, and calculates the energy distribution on the frequency axis at the parameter obtaining unit 4b. The parameter suitable for the external noise input or the parameter calculated in advance from the control table 6 is output from the data output unit 4c to the pre-processing unit 3. -19-200926138 In the processing unit 3, the frequency response control unit 3a and the gain control unit 3b are controlled based on the parameter 'obtained from the noise cancellation control unit 4', and the slave microphone 1 is passed through the microphone. The external noise input by the amplifier 2 is predicted by the "conduction function until the output means 1" function, and the output is "borrowed" by the reverse phase, thereby applying the noise canceling process. In this case, in the present invention, the device constituting the frequency response control unit 3a by the Q and the gain control unit 3b is an analog component that directly processes the external noise input as the analog signal as the state of the analog data. However, the amount of control is determined by the digital signal from the noise cancellation control unit 4. Therefore, according to the present embodiment, the cancellation processing of the external noise at the frequency response control unit 3a and the gain control unit 3b does not require A/D conversion or D/A conversion, and therefore has It is possible to minimize the delay of the noise cancellation processing. (4) Second Embodiment A second embodiment of the present invention will be described with reference to Fig. 6 . In the second embodiment, the reference microphone and the error microphone are used. In addition, the description of the same configuration as that of the first embodiment will be omitted. In Fig. 6, the 11-series is an error microphone, and the error microphone n is set to be a headset, a headphone, etc., for the sound actually heard by the listener (the sound that is eliminated by the noise). The output means -20- 200926138 ίο nearby. The error microphone signal from the error microphone 11 is amplified by the microphone amplifier 12, and then output to the post-processing unit 13 and the above-described noise canceling control unit 4. In this case, the error microphone 'signal is input to the post-processing unit 13 as an analog signal input from the error microphone 1 1 , and to the noise cancel control unit 4 via the A/D converter 14 . And as a digital data to be entered. The post-processing unit 13 includes a frequency response control unit 13a and a gain control unit 13b similarly to the pre-processing unit 3, and the control units 13a and 13b are eliminated by noise. The digital component of the control unit 4 is used as an analog component for control, and this is the same as the pre-processing unit 3 described above. In the second embodiment, the noise canceling control unit 4 generates the control unit 1 3 a for the post-processing unit 13 based on the error microphone signal in addition to the functions of the first embodiment. 1 3 b for control of the digital signal. In other words, the external noise detecting unit 4 includes a parameter acquiring unit 4b that acquires a specific parameter from the control table 6 based on the detection result of the external noise detecting unit 4a, and a parameter obtained by the parameter obtaining unit 4b. The control data is output to the frequency response control unit 3a and the data output unit 4c of the gain control unit 3b, and the frequency response control of the subsequent processing unit 13 is performed by the signal from the data output unit 4c. The unit 1 3 a is controlled by the gain control unit 1 3 b. However, in the noise control system which is typed after the second embodiment, the filter -15 - 200926138 filter 15 is disposed between the sound source and the D/A converter 9 thereof. In other words, in the error microphone signal recorded by the error microphone 11, since the residual portion of the cancelled noise and the music signal from the sound source 8 are present, in the post-processing unit 13, If the noise cancellation signal of the reverse phase is generated based on the error microphone signal, the music signal from the sound source will also be eliminated. Therefore, the correction filter 15 acquires the resource corresponding to the cancellation of the music signal at the post-processing unit 13 from the noise cancellation control unit 4, and responds to the source 8 from the sound source 8 The music signal is subjected to filtering processing, whereby even when the noise canceling signal corresponding to the music signal is applied from the post-processing unit 13, the original music signal can be not damaged. Correct the music signal from the sound source 8. In the noise control system which is employed after the second embodiment, it is necessary to update the correction filter 15 every time the parameter is updated. Therefore, in the second embodiment, the noise canceling φ control unit 4 is provided with the same filter 16 and the reverse phase generator 17, and the music signal from the output sound source 8 is taken as an example. The filter 15 is obtained in a later stage and compared with the error microphone signal from the error microphone 11, and the response to the sound source of the noise cancellation system is made using the same filter 16 (for example, an adaptive filter such as 'LMS). Similarly, in the inverse filter generator 17, the inverse filter is also approximated based on the same information. The correction filter 15 is updated by using the approximate inverse filter thus obtained, and can be pre-processed as if it were audible as if the original sound source signal was -22-200926138 as in the above, in the first In the second embodiment, the signal input from the error microphone 11 to the noise canceling control unit 4 is processed as a human body side of the earphone, that is, a sound actually heard in the ear. The mechanism that eliminates noise will be observed as the so-called "elimination of omissions." On the other hand, if attention is paid to the music signal from the sound source 8, it is possible to pass through the error microphone 1 1 and observe that "(the music that has been degraded by the noise cancellation processing) actually hears in the ear" . Further, in the noise canceling control unit 4, since the music signal of the reproduced sound is also input from the output side of the correction filter 15, it is possible to recognize that the music system is recognized by the sound of the error microphone Η. How has it deteriorated? And as the operation of the noise canceling control unit 4, the change amount of the signal is extracted from the original sound (music signal) and the error microphone signal by the same filter 16 of the LMS (from the speaker toward the system in the ear 〇 The impulse response)' and the output from the same filter 16 is used to generate an inverse filter in the inverse phase generator 17 (the conduction function when passing the aforementioned system is set to "1"). That is, the pulse response information of the above-described system is eliminated as a general filter, and the correction filter 15 is updated by this. As a result, according to the second embodiment, there is an advantage that the sound quality correction filter 15 can be changed in accordance with the change of the noise control parameter. Further, in the second embodiment, when the parameter of the digital control analog portion is limited, it is also possible to refer to the control table 6 based on the number of the reference -23-200926138 for the noise canceling unit. In the second embodiment, the external noise component of the error microphone signal detected by the error microphone 11 is input to the external noise detecting unit of the noise cancellation control unit 4 in the second embodiment. In 4a, the parameters of the control filter in the post-processing unit 13 are controlled. In this case, as in the foregoing, the information from the error microphone 11 is amplified by the microphone amplifier 12, and then digitally signaled by the A/D converter 14, which is "in the ear." The noise that should be heard directly is controlled. + The music signal to the ear." However, in the detection of external noise information, it is required to extract only the components of the music signal. Therefore, in the second embodiment, in the music signal from the digital sound source 8, the conduction function from the output means 10 to the error microphone 11 calculated in the item of the sound quality correction filter is made. The filter 18, which is the same as the convolution, is convoluted and subtracted from the error microphone signal which is converted into a digital signal by the A/D converter 14 in the adder 19. As a result, only the external noise component of the error microphone signal is input to the external noise detecting unit 4a, and can be the same as the external noise from the reference microphone 1, and the frequency of the subsequent processing unit 13 is formed. The parameters of the response control unit 1 3 a and the gain control unit 13b are determined. (5) In the third embodiment of the present invention, in order to show the third embodiment of the present invention, the parameter is determined based on the external noise obtained via the reference microphone in the first embodiment of the above -24-200926138, depending on the device. The characteristics of the individual or the characteristics of the individual's ear, and further in response to the environment in which the noise occurs, select the parameters desired by the user from among the plurality of parameters prepared and set in advance. In the third embodiment, the noise canceling control unit 4 is provided with the parameter selection switch 21 instead of the external noise detecting unit 4a, and is configured to be able to pass the parameter selection switch 21. The parameters obtained from the control table 6 of the parameter 0 acquisition unit 4b are selected. Therefore, in the third embodiment, the switch 21 is operated or adapted to the characteristics of the user's ear by the form or characteristics of the device such as the headphone or the earphone used. The switch 21 is operated to select the desired parameter, and the frequency response control unit 3a and the gain control unit 3b of the pre-processing unit 3 are controlled, and can be controlled. Perform appropriate noise cancellation processing. (6) Fourth Embodiment FIG. 8 shows a fourth embodiment of the present invention, and the fourth embodiment is a parameter selection switch 21 shown in the third embodiment. The pre-processing is combined with the subsequent processing shown in the second embodiment. According to the fourth embodiment of the above configuration, similarly to the second embodiment, the signal from the error microphone 11 can be used as a parameter, and the parameters of the noise state of the device or the surroundings can be adjusted. -25- 200926138 Decided. (7) In the fifth embodiment, the fifth embodiment of the present invention is shown. In the fifth embodiment, the reference microphone 1 is not used, and only the error microphone 11 is used to detect the external noise. , Correction of the sound quality caused by the correction filter 15. In addition, in FIG. 9, the parameter similar to that of the first embodiment is determined based on the external noise detected by the error microphone 11, but it may be similar to the third embodiment or the fourth embodiment. The parameter is determined via the parameter selection switch 21. In the fifth embodiment of the above configuration, similarly to the above-described respective embodiments, it is possible to determine the parameters of the surrounding noise state by giving back the signal from the error microphone 11. 〇 (8) Effects of the respective embodiments The effects of the respective embodiments having the above-described general configuration and function are as follows. In the above-described embodiment, in the noise canceling section which is required to have no delay response, only analog signal processing is used (digital control), and the delay is not a problem but the correct correction filter source is required. In the reproduction unit, only a digital signal processing system is applied. As a result, in the sound source reproducing unit, it is possible to incorporate a chopper including a non-causal element which is structurally impossible in the analog filter, and -26-200926138 In combination with the stereoscopic reproduction system in the headset, it is possible to construct a binaural system while performing noise cancellation. Further, by combining the correction filter unit with a digital signal and combining it with an adaptive filter, a correction filter for reproducing sound in the coefficient update of the noise canceling unit that requires complicated processing for maintaining the virtual system is realized. Update. In the above-described embodiment, the digital control system is incorporated in a noise canceling headphone system which is widely used, and the characteristic and gain of the system are made variable from the outside. It is characterized by the generation of a noise cancellation signal suitable for the condition of the device or the noise condition in the same system. In particular, by using analog components that can perform digital control, it is possible to realize a system that can perform digital control while having analogy responsiveness. In the foregoing embodiment, since the gain control system is a 1-block 1 parameter, and the frequency response control is also constituted by analog components, it is generally a parameter of about 1 or 2, and therefore, the number branch can also be used. The adaptation of the (Lap) LMS or the Perturbation methods is optimized. Therefore, it is possible to construct a kind of calculation that is much less than the noise cancellation with the widely-adjusted full-digital adaptive signal processing, and has the response characteristic that does not exist in the digital position. And available for practical systems. BRIEF DESCRIPTION OF THE DRAWINGS [Fig. 1] A block diagram of the first feature of the noise canceling system of the present invention is explained -27-200926138. Fig. 2 is a block diagram showing the second feature of the noise canceling system of the present invention. Fig. 3 is a block diagram showing the third feature of the noise canceling system of the present invention. Fig. 4 is a block diagram showing the fourth feature of the noise canceling system of the present invention. [Fig. 5] A block diagram showing a first embodiment of the noise canceling system of the present invention. Fig. 6 is a block diagram showing a second embodiment of the noise canceling system of the present invention. Fig. 7 is a block diagram showing a third embodiment of the noise canceling system of the present invention. Fig. 8 is a block diagram showing a fourth embodiment of the noise canceling system of the present invention. Fig. 9 is a block diagram showing a fifth embodiment of the noise canceling system of the present invention. Fig. 10 is a graph showing an example of a subjective evaluation curve (plot) and an example of an approximate curve in the present invention. Fig. 11 is a flow chart showing an example of a control system in the present invention. [Description of main component symbols] 1 : Reference microphone -28- 200926138 2 : Microphone amplifier 3 : Pre-processing unit 3 a : Frequency response control unit 3 b : Gain control unit 4 : Control unit 4a : External noise detecting unit 4b : Parameter acquisition unit Q 4c : Data output unit 5 : A/D converter 6 : Control table 7 : Adder 8 : Source 9 : D/A converter I 〇 : Output means II : Error microphone 〇 12 : Microphone amplifier 1 3 : post-processing unit 1 3 a : frequency response unit 1 3 b : gain control unit 14 : A/D converter Η : sound quality correction filter 1 6 : same filter 1 7 : inverse filter generator 18 : D/Α converter-29 200926138 1 9 : Adder 2 1 : Parameter selection switch
-30-30