201220299 六、發明說明： 【發明所屬之技術領域】 本發明關於一種音調偵測系統及聲音訊號調制方法。 【先前技術】 音調偵測系統一般用於偵測聲音中特定周期性的信號，例 如一音調。將原始聲音(audio source)插入多組特定頻率的音 調，這些音調的頻率分佈在人耳難以辨識的高頻區域(>15.8 KHz)且可組合成特定的指令，音調偵測系統偵測混合訊號中的 音調成分，之後即可根據音調組合成的特定指令作出相應的反 應，例如讓一玩具布偶作出手腳擺動、頭部晃動、發出聲音等 動作。 因為音調偵測系統是在一開放性的空間進行音調信號的 偵測，所以通常會遭遇如下問題。首先，聲音訊號在空間中傳 導會因環境四周的障礙物產生反射，如此反射訊號與原訊號到 達接收端時會產生時間差，而此時間差會產生回音(ech〇)導致 前後訊號彼此干擾。另外，聲音訊號在環境中傳導容易產生駐 波現象，當駐波產生時反射訊號與原訊號有接近反向的相位 差，導致聲音減弱或消失，大幅降低音調偵測的正確率。 [發明内容】 本發明提供-種具有良好音觸測正確率的音調彳貞測系統。 本發明的其他目的和優點可峨轉贿賊的脑特徵中得 刻進一步的了解。 依本發明-實施例之設計，一種音調_系統包含一接收器、— [S] 4 201220299 信號轉換裝置及-信號處理器。接收器接收對應—指令的一聲音信 ，’指令包含-起純元及減齡位元，聲音賴包含連續的複數 單位信號’該些單健號分麟應絲位元及指令位元且對應該些 指令位元的每一單位信號包含一段靜音期間及一單音調的一段持續 期間，其中兩相鄰該單位信號具有不同頻率的單音調。信號轉換= 置，用以將聲音信號轉換為包含單音調的—數位信號，且信號處理器 分析數位信號中的單音調以取得指令。 依本發明另-實補之設計種音調_系統包含—接收器、 一信號轉換裝置及一信號處理器。接收器接收對應一指令的一聲音信 號，指令包含一起始位元及複數指令位元，聲音信號包含連續的複數 單位信號，該些單位信號分別對應起始位元及指令位元，且對應指令 位元的每一單位信號至少包含一第一音調的一段持續期間、及—第二 音調的-段持續期間。第—音調的頻率大於第二音調的頻率，兩相鄰 單位信號具有相同頻率的第一音調及相同頻率的第二音調，第一音調 的該段持續期間與第二音調的該段持續期間相同。信號轉換裝置，用 以將聲音^雜換為包含帛—音麟第二音_—數位健，且信號 處理器分析數位信號中的第一音調及第二音調以取得指令。 依本發明另一實施例之設計，一種音調偵測系統包含一接收器、 一信號轉換裝置及一信號處理器。接收器接收對應一指令的一聲音信 號，指令包含一起始位元及複數指令位元，聲音信號包含連續的複數 單位信號，該些單位信號分別對應起始位元及指令位元，且對應指令 位元的每一單位信號至少包含一第一音調的一段持續期間、及一第二 音調的一段持續期間。第一音調的頻率大於第二音調的頻率，且兩相 鄰單位信號具有不同頻率的第一音調及不同頻率的第二音調。信號轉 換裝置用以將聲音信號轉換為包含第一音調及第二音調的一數位信 201220299 號，且信號處理ϋ分析數位信號中的第—音調及第二音調以取得指 〇 本發明另-實施例為—種聲音健_方法，鱗音信號係混入 於人耳可韻鱗音巾並職—齡，鱗音錄湖方法包含如下 步驟:接收該聲音信號及該人耳可賴鱗音，其中該指令包含一起 ，位/0及魏龄位元，鱗音健包含連細複料健號，該些 單位信號分麟應該起錄元及指令位元，且職雜指令位元 的每一該單位信號至少包含—段靜音_、—第—音調的—段持續期201220299 VI. Description of the Invention: [Technical Field] The present invention relates to a tone detection system and an audio signal modulation method. [Prior Art] A tone detection system is generally used to detect a specific periodic signal in a sound, such as a tone. The audio source is inserted into a plurality of sets of tones of a specific frequency whose frequencies are distributed in a high frequency region (> 15.8 KHz) that is difficult to recognize by the human ear and can be combined into a specific command, and the tone detection system detects the mixture. The tonal components in the signal can then be reacted according to the specific instructions that the tone is combined into, for example, a toy puppet swings the hands and feet, the head shakes, and the sound is emitted. Since the tone detection system performs tone signal detection in an open space, the following problems are usually encountered. First, the sound signal is transmitted in space and reflected by obstacles around the environment. When the reflected signal and the original signal arrive at the receiving end, a time difference occurs, and the time difference produces an echo (ech〇), causing the front and rear signals to interfere with each other. In addition, the sound signal is easily transmitted in the environment to generate a standing wave phenomenon. When the standing wave is generated, the reflected signal has a phase difference close to the original signal, which causes the sound to weaken or disappear, and the accuracy of the tone detection is greatly reduced. SUMMARY OF THE INVENTION The present invention provides a pitch measurement system having a good acoustic touch measurement accuracy rate. Other objects and advantages of the present invention provide further insight into the brain characteristics of the bribery. In accordance with the design of the present invention, a tone system includes a receiver, a [S] 4 201220299 signal conversion device, and a signal processor. The receiver receives a voice message corresponding to the command, the command contains - the pure element and the age-receiving bit, and the sound depends on the continuous complex unit signal. The single-sense number is divided into the bit bit and the command bit and Each unit signal of the instruction bits should contain a period of silence and a duration of a single tone, wherein two adjacent unit signals have a single tone of a different frequency. Signal Conversion = Set to convert the sound signal into a digital signal containing a single tone, and the signal processor analyzes the single tone in the digital signal to obtain the command. According to the invention, the design of the tone _ system includes a receiver, a signal conversion device and a signal processor. The receiver receives a sound signal corresponding to an instruction, the command includes a start bit and a plurality of command bits, and the sound signal comprises a continuous complex unit signal, wherein the unit signals respectively correspond to the start bit and the instruction bit, and the corresponding instruction Each unit signal of a bit includes at least a duration of a first tone, and a duration of a second tone. The frequency of the first tone is greater than the frequency of the second tone, the two adjacent unit signals have a first tone of the same frequency and a second tone of the same frequency, the duration of the segment of the first tone being the same as the duration of the segment of the second tone . The signal conversion device is configured to exchange the sound to include the second sound of the 帛-phone, and the signal processor analyzes the first tone and the second tone of the digital signal to obtain an instruction. According to another embodiment of the present invention, a tone detection system includes a receiver, a signal conversion device, and a signal processor. The receiver receives a sound signal corresponding to an instruction, the command includes a start bit and a plurality of command bits, and the sound signal comprises a continuous complex unit signal, wherein the unit signals respectively correspond to the start bit and the instruction bit, and the corresponding instruction Each unit signal of the bit includes at least a duration of a first tone and a duration of a second tone. The frequency of the first tone is greater than the frequency of the second tone, and the two adjacent unit signals have a first tone of a different frequency and a second tone of a different frequency. The signal conversion device is configured to convert the sound signal into a digital letter 201220299 including the first tone and the second tone, and the signal processing/analyzes the first tone and the second tone in the digital signal to obtain the fingerprint. For example, the sound sound _ method, the scalar sound signal is mixed into the human ear can be scaled sound towel, and the age, the scale sound recording lake method comprises the following steps: receiving the sound signal and the human ear can be scaled sound, wherein the instruction Including together, bit / 0 and Wei Ling bit, scale sound health contains a fine resurrection health, these unit signals should be recorded from the recording element and the instruction bit, and each unit signal of the miscellaneous command bit Contain at least - segment mute _, - first - tone - segment duration
間、及-第二音調的_段持續綱’該第—音調的頻率大於該第二音 調的頻率’且__單位信號具有不同鮮的該第—音調及不同頻 ，的該第—a調’賴聲音錢轉換為—數位錢；及分析該數位信 號以取得該指令。 發月另Μ施例為-種聲音信號調制方法，該聲音信號係混入 於人耳可顺鱗音巾並職，縣音錢卿方純含如下 步驟4收該聲音域及該人耳可賴鱗音，其中職令包含-起 ，位几及複數指令位元’該聲音錄包含連_魏單健號該些 1信號分麟鏡起始位元及該些指令位元，且對應該些指令位元 1每一該^位舰至少包含-段靜音躺、—第—音調的—段持續期 第二音調的一段持續期間，其令該第一音調的頻率大於該第 頻率率兩相籠單位錢具有烟辭的娜-音調及相同 持鋒期1，該第—音調的該段持續綱與該第二音調的該段 _，且該段靜音_的長度大於顧持續朗的長度的 卞將鱗音減轉縣—數健號·及分析魏健號以 該指令。 包含一訊號放大器、一 本發明另-實施例為—種音調偵測系統 201220299 濾波器、一類比數位轉換器及一數位信號處理器。訊號放大器接收並 放大對應一指令的一聲音信號，其中該指令包含一起始位元及複數指 令位元，聲音信號包含連續的複數單位信號，該些單位信號分別對應 起始位元及指令位元’該些單位信號分屬不同頻帶且各該單位信號包 含持續-段時間的至少-音調。濾波减除聲音信號的雜訊，類比數 位轉換器用轉聲音健轉麟包含指令的—數位㈣^數位信號處 理器分析數位信號以取得指令，其中分別屬於不同頻帶的複數個音調 的頻率值彼此之間具有一預設的加權關係。 藉由上述各個實施例之設計，當兩個相鄰的單位訊號的音調頻率 不同時，可有效減少前後訊號干擾的回音現象，且當每一單位信號均 為同時包含複數音調的多音調訊號時，可有效降低因聲音駐波現象造 成的錯誤率。 本發明的其他目的和優點可峨本發騎财的技術特徵中得 到進-步的了解。為讓本發明之上述和其他目的、特徵和優點能更明 顯易懂，下文特舉實施例並配合所附圖式，作詳細說明如下。 【實施方式】 有關本發明之前述及其他技躺容、_與功效，在町配合參 考圖式之實施例的詳細說财，將可清楚的呈現。以τ實施例中所提 到的方向聽’例如：上、τ、左、右、前或後等，僅是參考附加圖 式的方向。·，的方_語是用來說縣_來限制本發明。 如圖1所示，音調偵測系統10可接收對應一指令的一聲 音信號並使-反應裝置3〇產生—反應，該聲音信號例如可混 入人耳可觸的聲音巾並包含分佈在人耳㈣韻的高頻區 域(>15.8 KHz)的至少-音調’且該人耳可辨識的聲音及該聲音 201220299Between the second and the second syllables, the frequency of the first tone is greater than the frequency of the second tone, and the __ unit signal has different first and second tones of the first tone and the different frequencies. 'Lai voice money is converted to - digital money; and the digital signal is analyzed to obtain the instruction. Another example is the sound signal modulation method. The sound signal is mixed into the human ear and can be used in conjunction with the scales. The county voice Qian Qingfang contains the following step 4 to receive the sound field and the human ear. Where the order contains - from, a few bits and a plurality of command bits 'The sound record contains the _ Wei Danjian number 1 signal splitting mirror start bit and the instruction bit, and corresponding to some instruction bit 1 Each of the ships includes at least a period of silence, a period of the first tone of the second tone, and the frequency of the first tone is greater than the frequency of the first phase. The Na-tone of the vocabulary and the same holding period 1, the segment of the first-tone is continued with the segment _ of the second tone, and the length of the segment mute _ is greater than the length of the continuation lang. Turn the county - number health number and analyze the Wei Jian number with this instruction. A signal amplifier is included, and another embodiment of the invention is a tone detection system 201220299 filter, an analog-to-digital converter and a digital signal processor. The signal amplifier receives and amplifies a sound signal corresponding to an instruction, wherein the command includes a start bit and a plurality of command bits, and the sound signal comprises a continuous complex unit signal, wherein the unit signals respectively correspond to the start bit and the instruction bit The unit signals belong to different frequency bands and each of the unit signals contains at least a tone of continuous-segment time. Filtering and subtracting the noise of the sound signal, the analog digital converter analyzes the digital signal by using a digital-to-digital (four) digital signal processor containing instructions, wherein the frequency values of the plurality of tones belonging to different frequency bands are mutually There is a preset weighting relationship between them. According to the design of the foregoing embodiments, when the pitch frequencies of two adjacent unit signals are different, the echo phenomenon of front and rear signal interference can be effectively reduced, and when each unit signal is a multi-tone signal including multiple tones at the same time. , can effectively reduce the error rate caused by the sound standing wave phenomenon. Other objects and advantages of the present invention will provide further insight into the technical features of the present invention. The above and other objects, features, and advantages of the present invention will become more fully understood from [Embodiment] The detailed description of the embodiments of the present invention in conjunction with the embodiments of the present invention will be apparent. Listening in the direction mentioned in the τ embodiment, for example: upper, τ, left, right, front or back, etc., is only the direction in which the additional pattern is referred. · The _ _ language is used to limit the invention. As shown in FIG. 1, the tone detection system 10 can receive a sound signal corresponding to an instruction and cause a reaction to the reaction device 3, which can be mixed into a human ear and can be included in the human ear. (d) at least the tone of the high frequency region of the rhyme (> 15.8 KHz) and the audible sound of the human ear and the sound 201220299
L號同時Φ播放裝置20發出。麥克風M接收播放裝置2〇 發出的人耳可韻的聲音及聲音錢並傳送至—音調偵測系 統^音議測系統H)包含-接收器12、—信號轉換裝置14及 仏號處理器16。接收器12接收該聲音信號，信號轉換裝置 Η用以將聲音錢㈣為包含音__數位信號，信號處理器 16分析數位信號中的音調以取得指令1應裝置3()接收指令 並產生-反應’該反應舉例而言可為—玩具布偶的手腳擺動、 頭部晃動、發出聲音等動作。於_實施例中，接收器12可包 含-訊號放大器及-舰器，以放大聲音信號並雜訊。信 號轉換裝置14例如可為—類比數轉換器㈣士州以咖 C〇nVerter ; ADC)，且信號處理器16例如可包含-數位信號處 理器（digital signal process〇r; Dsp)。數位信號處理器可即時分 析數位信號中的音調以取得指令。 於-實施例中，指令包含一起始位元及複數指令位元，聲 音信號包含連續的複數單位㈣’各個單位信號分別對應該起 始位元及該些指令位元，且對應該些指令位元的每_單位信號 至少包含-段靜音腳、—高音調的_段持續㈣、及一低音 調的-段持續期間。舉例而言，如圖2所示，指令位元可包含 -起始位元IMS指令位元(位元〇·位元7)，且制起始位元 及指令位元的單位信號組成舉例而言如下表所示：The L number is simultaneously emitted by the Φ playback device 20. The microphone M receives the sound and sound money of the human ear that is emitted by the playing device 2 and transmits it to the tone detecting system (the tone detecting system H) including the receiver 12, the signal converting device 14 and the nickname processor 16 . The receiver 12 receives the sound signal, the signal conversion device is configured to use the sound money (4) as a sound-containing digital signal, and the signal processor 16 analyzes the tone in the digital signal to obtain the command 1 to receive the command from the device 3 () and generate - The reaction 'exemplary' can be, for example, an action of the toy puppet swinging, shaking the head, making a sound, and the like. In the embodiment, the receiver 12 may include a signal amplifier and a carrier to amplify the sound signal and the noise. The signal conversion device 14 can be, for example, an analog-to-digital converter (4), or a digital signal process (Dsp). The digital signal processor instantly analyzes the tones in the digital signal to obtain instructions. In the embodiment, the instruction includes a start bit and a complex instruction bit, and the sound signal includes consecutive complex units (four) 'each unit signal respectively corresponding to the start bit and the instruction bits, and corresponding to the instruction bits Each _ unit signal of the meta includes at least a segment silent leg, a high pitch _ segment continuous (four), and a low tone segment - segment duration. For example, as shown in FIG. 2, the instruction bit may include a start bit IMS instruction bit (bit 〇·bit 7), and the unit signal composition of the start bit and the instruction bit is exemplified. The words are as follows:
位元 7 :邏輯值 1 Ϋΐ(〇·〇〇5 #T^Wc(0^5i!〇 201220299 (頻帶1) d(〇.〇2?#) -- 位元6:邏輯值1 (頻帶2) 靜音(0.005秒)+音調g(〇.025秒） +音調H(0.025秒） 位元5 :邏輯值〇 (頻帶1) 靜音(0.005秒)+音調Ε(0·025秒） +音調F(0.025秒） 位元4 :邏輯值〇 (頻帶2) 靜音(0_005秒)+音調1(0.025秒） +音調J(0.025秒） 位元3 :邏輯值1 (頻帶1) 靜音(0·005秒)+音調C(0.025秒） +音調D(0.025秒） 位元2 :邏輯值1 (頻帶2) 靜音(0.005秒)+音調G(〇.〇25秒） +音調H(0.025秒） 位元1 :邏輯值0 (頻帶1) 靜音(0.005秒)+音調e(〇.〇25秒） +音調F(0.025秒） 位元0 :邏輯值〇 (頻帶2) 靜音(0.005秒)+音調1(0.025秒） +音調J(0_025秒） 如上表所示並請同時參考圖3的聲音訊號時域圖，於本實例 中，每一單位信號均為同時包含兩個音調的雙音調訊號，例如 對應起始位元的單位信號包含持續一段時間的一高音調及一 低音調’且對應指令位元的每—單位信號包含_段靜音期間、 -高音調的-段持續期間、及—低音調的—段持續期間。舉例 而言，對應起始位元的單位信號為持續〇 〇4秒的音調A及音 調B，對應位元7的單位信號包含持續〇 〇〇5秒的一段靜音及 持續0.025秒的音調C及音調d。 因本實施例的每-單位信號均為同時包含兩個音調的雙 [S1 9 201220299 音調訊號，亦即同一時間點存在兩個單音調信號，如此在同一 時間點兩個單音調彳§號都因駐波而消失的機率相對降低，故多 音調訊號的設計（不限定為雙音調）可有效降低因聲音駐波現象 造成的錯誤率。再者，於本實施例中，兩相鄰單位信號彼此的 高音調頻率及低音調頻率均不相同。如圖3所示，舉例而言， 對應位元7的高音調C與相鄰對應位元6的高音調G兩者的頻 率不同，且對應位元7的低音調D與相鄰對應位元6的低音調 Η兩者的頻率不同。一般而言，聲音訊號在空間傳導會因環境 四周的障礙物產生反射，如此反射訊號與原訊號到達接收端時 會產生時間差，而此時間差會產生回音(echo)導致前後訊號彼 此干擾。因本實施例設計使兩個相鄰的單位訊號的音調頻率不 同，因此可有效減刖後訊號干擾的回音現象。於一實施例 中’音調A至音調J的分布範圍為15.8kHz_2〇kHz，且兩相鄰 單位訊號的頻率差異值較佳為大於100Hz。另外，於一實施例 中，對應起始位元的單位信號的高音調頻率可小於對應指令位 元的事位信號的南音調的最低頻率’或者對應起始位元的單位 信號的低音調頻率可大於對應指令位元的單位信號的低音調 的最高頻率。 於一實施例，若要增加單位信號間的緩衝空間且不増加指 令信號的長度，可使用更多的頻帶來組成一個指令，例如圖4 所示，可使用三個頻帶(頻帶1、頻帶2、頻帶3)來組成—個指 令，且一個指令包含9個指令位元。因圖4兩個相鄰頻帶丨（中 間間隔頻帶2、頻帶3)的緩衝空間會大於圖3兩個相鄰頻帶中 間僅間隔頻帶2)的緩衝空間，所以圖4的單位信號的長度τ, 可設為小於圖3的單位信號的長度Τ。 IS] 10 201220299 圖5為顯示本發明另一實施例的聲音信號時域圖。如圖5 所不，對應指令位元的每一單位信號至少包含一段靜音期間 TS、一高音調C的一段持續期間Td、及一低音調D的一段相 同持續期間Td。本實施例的兩相鄰單位信號具有相同頻率的高 音調C及相同頻率的低音調D，且靜音期間的長度Ts設為大 於音調持_間Td的長度的μ倍4然本實關的兩相鄰單 位信號具有相同頻率，但因靜音期間的長度丁3設為大於音調持 續期間Td的長度的M倍’所⑽提供足夠的緩衝空間避免同 頻率的訊號因反射而相互干擾，在4坪以上的封閉空間可獲得 95%以上的音調偵測正確率。 再者，依本發明之設計，指令位元數目可依需求調整，且 對應起始位元的單位信號及對應指令位元的單位信號的長度 均可視需求調整，例如圖6所示，兩相鄰的單位信號可具有不 同的長度(T1>T2)或不同的音調持續時間(Tdl<Td2)，且單位信 號亦可為單音調信號，僅需當兩相鄰單位信號的音調頻率不同 時，即可有效減少前後訊號干擾的回音現象。 再者，本發明提供一種聲音信號調制方法，該聲音信號係 混入於人耳可辨識的聲音中並對應一指令。聲音信號調制方法 例如包含如下步驟：利用一接收器同時接收聲音信號及人耳可 辨識的聲音；將聲音信號轉換為一數位信號；及分析數位信號 以取得指令。指令的内容則如前述各個實施例所示。 另外，於一實施例中，上述之接收器12、信號轉換裝置 14及信號處理器16整合於一積體電路晶片。 圖7為依本發明另一實施例’顯示於音調中加入校正碼方 式的示意圖。如圖7所示，指令是由音調s、S卜音調A、A1、 m 11 201220299 音調B、B1和音調C、C1所組成，音調S、S1是起始單位訊 说’音調A、A1、音調B、B1和音調C、Cl分別為頻帶1、 頻帶2和頻帶3的單位信號。於本實施例中，各個頻帶的音調 彼此之間需符合一預設的關係式’該些單位信號才會被認定為 包含音調偵測系統1 〇的編碼指令。於一實施例中，以單一^通 道為例可符合如下關係式：Bit 7: Logic value 1 Ϋΐ(〇·〇〇5 #T^Wc(0^5i!〇201220299 (Band 1) d(〇.〇2?#) -- Bit 6: Logic value 1 (Band 2 ) Mute (0.005 sec) + Tone g (〇.025 sec) + Tone H (0.025 sec) Bit 5: Logic value 〇 (Band 1) Mute (0.005 sec) + Tone 0 (0·025 sec) + Tone F (0.025 sec) Bit 4: Logic value 〇 (Band 2) Mute (0_005 sec) + Tone 1 (0.025 sec) + Tone J (0.025 sec) Byte 3: Logic value 1 (Band 1) Mute (0·005) Second) + Tone C (0.025 sec) + Tone D (0.025 sec) Bit 2: Logic 1 (Band 2) Mute (0.005 sec) + Tone G (〇. 〇 25 sec) + Tone H (0.025 sec) Element 1: Logic value 0 (Band 1) Mute (0.005 sec) + Tone e (〇.〇25 seconds) + Tone F (0.025 sec) Bit 0: Logic value 〇 (Band 2) Mute (0.005 sec) + Tone 1 (0.025 sec) + tone J (0_025 sec) As shown in the above table, please also refer to the sound signal time domain diagram of Figure 3. In this example, each unit signal is a two-tone signal containing two tones at the same time. For example, the unit signal corresponding to the start bit includes a high pitch and a low tone for a period of time and corresponding commands Each unit signal of the element includes a period of _ segment silence, a period of high pitch - duration of the segment, and a duration of - low pitch. For example, the unit signal corresponding to the start bit is 〇〇 4 seconds. Tone A and tone B, the unit signal of the corresponding bit 7 contains a period of silence of 5 seconds and a tone C and tone d of 0.025 seconds. Since each unit signal of the embodiment includes two tones simultaneously. The dual [S1 9 201220299 tone signal, that is, there are two single tone signals at the same time point, so that the probability of the two single tone 彳 § signs disappearing due to the standing wave at the same time point is relatively reduced, so the design of the multi-tone signal (not limited to double tone) can effectively reduce the error rate caused by the sound standing wave phenomenon. Moreover, in this embodiment, the high pitch frequency and the low tone frequency of the two adjacent unit signals are different from each other. As shown, for example, the high pitch C of the corresponding bit 7 is different from the high pitch G of the adjacent corresponding bit 6, and the low pitch D of the corresponding bit 7 and the bass of the adjacent corresponding bit 6 are shown. Tune the frequency of the two In general, the sound signal is reflected in the space due to obstacles around the environment. When the reflected signal and the original signal arrive at the receiving end, a time difference will occur, and the time difference will cause an echo to cause the front and rear signals to interfere with each other. The embodiment is designed to make the tone frequencies of two adjacent unit signals different, so that the echo phenomenon of the signal interference can be effectively reduced. In one embodiment, the distribution of pitch A to pitch J is 15.8 kHz 2 〇 kHz, and the frequency difference value of two adjacent unit signals is preferably greater than 100 Hz. In addition, in an embodiment, the high pitch frequency of the unit signal corresponding to the start bit may be smaller than the lowest frequency of the south tone of the event signal corresponding to the instruction bit or the low frequency of the unit signal corresponding to the start bit. It may be greater than the highest frequency of the low tone of the unit signal corresponding to the command bit. In an embodiment, if the buffer space between the unit signals is to be increased and the length of the command signal is not increased, more frequency bands may be used to form an instruction. For example, as shown in FIG. 4, three frequency bands (band 1, band 2) may be used. Band 3) is composed of one instruction, and one instruction contains 9 instruction bits. Since the buffer space of two adjacent frequency bands 丨 (intermediate interval frequency band 2, frequency band 3) in FIG. 4 is larger than the buffer space of only two adjacent frequency bands 2 in the middle of FIG. 3, the length τ of the unit signal of FIG. 4, It can be set to be smaller than the length 单位 of the unit signal of FIG. IS] 10 201220299 FIG. 5 is a time-domain diagram showing a sound signal according to another embodiment of the present invention. As shown in Fig. 5, each unit signal corresponding to the command bit includes at least one period of silence TS, a period of duration Td of a high pitch C, and a period of the same duration Td of a low tone D. The two adjacent unit signals of the embodiment have a high-pitched tone C of the same frequency and a low-pitched tone D of the same frequency, and the length Ts of the silent period is set to be larger than the μ times of the length of the tone-holding Td, and the two are actually closed. The adjacent unit signals have the same frequency, but because the length of the mute period is set to be M times greater than the length of the tone duration Td (10), sufficient buffer space is provided to prevent the signals of the same frequency from interfering with each other due to reflection, at 4 pings. The above enclosed space can achieve more than 95% tone detection accuracy. Furthermore, according to the design of the present invention, the number of instruction bits can be adjusted according to requirements, and the unit signal corresponding to the start bit and the length of the unit signal corresponding to the instruction bit can be adjusted according to requirements, for example, as shown in FIG. The adjacent unit signals may have different lengths (T1>T2) or different tone durations (Tdl<Td2), and the unit signal may also be a single tone signal, only when the pitch frequencies of two adjacent unit signals are different. It can effectively reduce the echo of front and rear signal interference. Furthermore, the present invention provides a method of modulating a sound signal that is incorporated into a recognizable sound of a human ear and corresponds to an instruction. The sound signal modulation method includes, for example, the steps of: simultaneously receiving a sound signal and a sound recognizable by the human ear by using a receiver; converting the sound signal into a digital signal; and analyzing the digital signal to obtain an instruction. The contents of the instructions are as shown in the various embodiments described above. In addition, in an embodiment, the receiver 12, the signal conversion device 14, and the signal processor 16 are integrated in an integrated circuit chip. Figure 7 is a diagram showing the addition of a correction code pattern to a tone in accordance with another embodiment of the present invention. As shown in Fig. 7, the command is composed of pitch s, S-pitch A, A1, m 11 201220299 tone B, B1 and pitch C, C1, and the pitches S and S1 are the initial unit voices 'tones A, A1. Tones B, B1 and tones C, Cl are unit signals of Band 1, Band 2 and Band 3, respectively. In this embodiment, the tones of the respective frequency bands are required to conform to a predetermined relationship with each other. The unit signals are recognized as encoding instructions including the tone detecting system 1 . In an embodiment, a single channel can be used as an example to satisfy the following relationship:
W, X ToneA + W2 x ToneB = W3 x ToneC ’其中ToneA、ToneB、ToneC分別代表音調a、音調B、音調 C的頻率值，且Wi、W2和W3分別表示於同一通道中的各個 音調的權重。當分別屬於不同頻帶的複數個音調的頻率值 ToneA、ToneB、ToneC，彼此之間具有例如上述的預設加權關 係時，數位信號處理器(DSP)即可解碼辨識出聲音信號中包含 音調偵測系統10要求的編碼指令。 雖然本發明已讀佳實施觸露如上，然其並非载限定本發 明，任何熟習此技藝者，在不脫離本發明之精神和範圍内，當可作些 許之更動_飾，S此本發明之保護翻當視後社_請專利範圍所 界定者鱗。另外，本發_任—實施贼巾請專利劍獨達成本 發明所揭露之全部目的紐點或獅。此外，摘要部分和標題僅是用 來輔助專敝件搜尋之S，麟用來闕本發明之權利範圍。 【圖式簡單說明】 圖1為顯示本發明一實施例的音調制系統的方塊圖。 圖2為顯示本發明一實施例的指令組成示意圖。 圖3為顯示本發明一實施例的聲音信號時域圖。 [S] 12 201220299 圖4為顯示本發明另一實施例的聲音信號時域圖。 圖5為顯示本發明另一實施例的聲音信號時域圖。 圖6為顯示本發明另一實施例的聲音信號時域圖。 圖7為顯示本發明另一實施例的音調加入校正碼的示意圖。 t主要元件符號說明】 1〇 音調偵測系統 A-L 音調 12 接收器 Μ 麥克風 14 信號轉換裝置 Τ、Τ’、Ή、Τ2單位信號長度 16 信號處理器 Td、 Tdl、Td2音調持續期間 2〇 播放裝置 Ts 靜音期間 3〇 反應裝置W, X ToneA + W2 x ToneB = W3 x ToneC 'where ToneA, ToneB, ToneC represent the frequency values of tone a, tone B, and tone C, respectively, and Wi, W2, and W3 represent the weights of the respective tones in the same channel, respectively. . When the frequency values ToneA, ToneB, and ToneC of the plurality of tones respectively belonging to different frequency bands have a predetermined weighting relationship with each other, for example, the digital signal processor (DSP) can decode and recognize the tone detection included in the sound signal. The encoding instructions required by system 10. Although the present invention has been described as a preferred embodiment, it is not intended to limit the invention, and those skilled in the art can make a few modifications without departing from the spirit and scope of the invention. Protection is turned into a post-consultation _ please define the scale of the patent scope. In addition, this hair _ 任 - implementation of thief towel, please patent sword to achieve the full purpose of the invention disclosed in the point or lion. In addition, the abstract sections and headings are only used to assist the exclusive search of S, which is used to claim the scope of the invention. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a sound modulation system according to an embodiment of the present invention. 2 is a schematic diagram showing the composition of an instruction according to an embodiment of the present invention. Fig. 3 is a timing chart showing a sound signal according to an embodiment of the present invention. [S] 12 201220299 FIG. 4 is a time-domain diagram showing a sound signal according to another embodiment of the present invention. Fig. 5 is a time-domain diagram showing a sound signal according to another embodiment of the present invention. Figure 6 is a time-domain diagram showing a sound signal of another embodiment of the present invention. Figure 7 is a diagram showing a pitch addition correction code according to another embodiment of the present invention. t main component symbol description] 1〇 tone detection system AL tone 12 receiver 麦克风 microphone 14 signal conversion device Τ, Τ ', Ή, Τ 2 unit signal length 16 signal processor Td, Tdl, Td2 tone duration 2 〇 playback device Ts mute period 3〇 reaction device