五、發明說明（【發明領域】 本發明是有關於—^ 、種曰效處理方法及裝置，特別是指 一種利用調整取樣頻率（ ^ 貝丰（沾师11叫rate)來降低音效處理所 需之s己憶體容量之麥克風用 凡风用的音效處理方法及裝置。5【習知技藝說明】 I麥克風音效處理大致可分成類比與數位訊號的兩 種音效處理方式。目前相較於類比音效處理方式，數位音 效處理方式能獲得更佳的音效效果。 由於麥克風1係將為聲波的訊號轉換類比電訊號以及 揚聲器2係將類比電訊號轉換成聲波輸出，所以數位音效 處理裝置3内需將自麥克風！之訊號數位化後再進行音效 處理，而在音效處理後需再將訊號類比化後以適於供揚聲 器2輸出。因此，如第一圖，係一種習用麥克風用的數位 音效處理裝置3之方塊圖。此裝置3包含一接收麥克風i 輸出之類比訊號並將其數位化之類比至數位轉換器 (analog-to-digital converter，簡稱 ADC)31、一 暫存數 位化的語音訊號之記憶體32、一自記憶體32擷取語音訊號 進行訊號處理之數位訊號處理器（digital signal processor，以下簡稱DSP)33、以及一接收經DSP33處理訊 號並將其類比化以適於傳送至揚聲器2之數位至類比轉換 為（digital-to-analog converter，簡稱 DAC)34。藉此， 當麥克風1將聲波轉換成類比電訊號時，則可經ADC31先 將訊號數位化儲存至記憶體32，而後DSP33依需要自記憶 體32讀取訊號後並加以處理以作成各種效果，例如回音 10 15 20 第4頁 --------^--------- (請先閱讀背面之注意事項再填寫本頁) % 家標準(CNSM4 規格（210x 297公釐） 532045 A7 經 濟 部 智 慧 財 產 局 消 費 合 社 印 製 五、發明說明（2 (eCh〇)、合音（chorus)、顫音（flanger)、升降音（pitch Shlft)、等化器（equalizer)、變聲、壓縮及解壓縮、語音 =識等等，其後再由DAC34將訊號轉換成類比訊號，以: 揚聲器2播放。 /、 、、近年來隨著多媒體娛樂的興起，讓使用者對於麥克風 要求的9質亦隨之提高，而影響音效處理後音質的其中一 f要因素為取樣頻率。此取樣頻率意指在類比至數i立轉換 為31將類比訊號轉換成數位訊號過程中，以週期性的固定 頻率榻取連續資料的技術。一般來說，取樣頻率需高於音 源（對麥克風來說即為人聲）之頻率，而隨著取樣頻率兪 則音質愈好。 今曰電子零件趨向於整合與微小化的趨勢，使得習用 的音效處理裝置3亦會以系統整合於單一積體電路被形 成，甚者可整合於單一晶片上。但隨著取樣頻率的增加， 使得音效處理裝置3中的記憶體32容量與對應的Dsp所需 的 MIPS(milllon instructi〇ns 咐 se_d;每秒百萬個 指令）量亦隨之大幅增加，舉例來說，諸如回音之音效處理 率與各取樣所需的儲存位元（blt)所決以記憶體容量=最 =延遲時間*取樣頻率*各取樣之儲存位元），假定最大延遲 時間為300ms、取樣頻率為四萬八千赫兹（Hz)以及各取樣 之儲存位兀為16位元，則記憶體32容量就需28_位元 組（即0.3*·續6=23咖blts=2咖bytes)。如此，音效處 理裝置1要求如此高的記憶體32的容量與删3的運算能 5 10 訂 15 線 20 五、發明說明（3 ) " "— -- 力，因此需佔用過多的電路面積，造成習用的音效處 置3較難整合，甚至成為整合於_晶片上的重大負擔。、 若麥克風1的聲音不經任何音效處理即傳送至揚聲器 2播放，則取樣頻率應盡可能地高，進而以較佳音質播放。 然而，實際上取樣速度愈快，則人類愈無法察覺兩次取樣 之間的差異性’而且由於聲音經音效處理原本就會變化， 舉例來說，回音是模擬聲音在大空間下經由牆壁反彈後回 傳的殘響，這樣的回音經牆壁與空氣的吸收衰減’多半本 身音質就不佳，所以人們對於經過音效處理的聲音反而不 企求像聲音直接輸出的高音質。以麥克風i應用來說，取 樣頻率約為八千赫兹對於人的聲音來說已經棒縛有餘。因 此在不影響麥克風輸出音質下，若降低音效處理之取樣頻 率’隨之記憶體32容量與DSP33運算能力之要求亦可隨之 降低，進而解決以往存在的問題。 1 5 【發明概要】 因此，本發明之-目的，乃在提供一種利用於音效處 理時調降取樣頻率以節省記憶體容量與降低對運算能力要 求之麥克風用的音效處理方法。 % *發明之另—目的，乃在提供—種麥克風用的音效處 ▲理方法，係利用於音效處理時調降取樣頻率並於音效處理 後恢復成原本取樣頻率以與原本未經音效處㈣號混音， 以達到在維持高音質下降低音效處理所需記憶體容量之功 效。 本發明之再-目的，乃在提供—種可達到降低音效處 532045 A7 工 五、發明說明（4 ) 理所需的記憶體容量之麥克風用的音效處理裝置。 本[月之t目的’乃在提供-種可達到降低對運算 月匕力要求之麥克風用的音效處理裝置。 於是，本發明之麥克風的音效處理方法，係適於音效 處理經-麥克風掏取之語音訊號後傳送至一揚聲器播放， 該方法包含以下步驟·· ! —)將自》麥克風之語音訊號以為一第一頻率之取樣頻 ^ 率數位化成一原始語音訊號後輸出； )將v驟A )中的5亥原始語音訊號訊號之取樣頻率調降 後進行音效處理，其後，再將經處理的訊號之取樣頻率恢 復成該第一頻率，以形成一音效訊號；以及 C)接收該步驟a)之原始語音訊號與該步驟β)之該音 效訊號並混音成-聲音訊號，其後更將聲音訊號類比化以 適於輸出至該揚聲器轉換成聲波播放。 再者，本發明之音效處理裝置包含一類比至數位轉換 器、一音效處理單兀、一混音器及一數位至類比轉換器· 其中，該類比至數位轉換器係接收該麥克風之語音訊號 係以第一頻率作為取樣頻率數位化該語音訊號形成一 始語音訊號；該音效處理單元係自該類比至數位轉換器状 收忒原始浯音訊號並調降該原始語音訊號之取樣頻率來進 行音效處理及在音效處理恢復該語音訊號之取樣頻率， 形成一音效汛號；該混音器係分別自該類比至數位轉換w 與該音效處理單元接收該原始語音訊號與該音效訊號後合 成一聲音訊號；及該數位至類比轉換器係自該混音器接收 請 5 10 訂 15 並 原 接 線 20 以 器 第7頁 本紙張尺度適用中國國家標準(CNS)A4規格（210 X 297公餐） 532045 15 20 Λ7 五、發明說明（5 ) 該聲音訊號並類比化該聲音訊號以適於輪出至該揚聲器播 【圖式之簡單說明】 本發明之其他特徵及優點，在以下配合參考圖式之較 5佳實施例的詳細說明中，將可清楚的明白，在圖弋中· 乂 第一圖是習用麥克風用的數位音效處理裝置的電路方 塊圖，此裝置是適於接收-麥克風之訊號並進行處理後適 於輸出至一揚聲器播放； 第二圖是本發明之較佳實施例之電路方塊圖，此實施 10 例係結合一麥克風與一揚聲器；及 第二圖是第二圖中較佳實施例之流程圖。 【較佳實施例之詳細說明】 參閱第二圖，是顯示本發明之麥克風用的音效處理裝 置6之一較佳實施例。此音效處理裝置6係適於接收一麥 克風4擷取之類比聲音訊號並經處理後適於傳送至一揚聲 為5播放。本實施例之音效處理裝置6係包含一類比至數 位轉換器（ADC)6卜降頻單元62、一記憶體63、一數位訊 唬處理單tc(DSP)64、昇頻單元65、一混音器66及一數位 至類比轉換器67。 此ADC61係電性連接至麥克風4、降頻單元62與混音 器66。在本實施例中，麥克風4轉換為聲波之語音成一類 比語音訊號，而ADC61以為一第一頻率之取樣頻率（例如 48kHz)將麥克風4輸出之類比語音訊號轉換成數位原始語 音訊號後分別輪出至降頻單元62與混音器66。 第8頁 本紙張尺度翻㈣目^ 事 f 532045 Λ7 B7 五、發明說明（6 10 此降頻單元62用以調降自ΑηΓβ1 ^ 门降自ADC61之原始語音訊號之苹 ::!成一低於第一頻率之第二頻率後輸出至記憶體63。 ί 例中，由於音效處理m係供麥克風4應用， 故弟-取樣頻率僅足以應付人類的聲音即可，換言之，第 -頻率僅需至少等於八千赫兹即可，舉例來說第二頻率為 :二萬四赫兹或一萬二赫茲等等，而依設計者需求來決定 第二頻率的數值。 此記憶體63用以暫存經降頻單元62降頻至第二頻率 之語音訊號，以供後續的DSP64自記憶體63内擷取語音訊 號進行適當處理並且可供DSP64於處理語音訊號過程中資 料存取。 、 . ^ — (請先閱讀背面之注意事項再填寫本頁) 5 ο 1 2_i_ 經濟部智慧財產局員工消費合作社印製 此DSP64係電性連接記憶體63與昇頻單元64並具有 語音訊號處理功能，以用來自記憶體63中擷取數位語音訊 號依所品g效進行机號資料處理，而音效處理包含回音、 &曰頦9、升降音、等化器、變聲、壓縮及解壓縮、語 音辨識等等。由於DSP64進行語音訊號處理時所需的記憶 體會因DSP64進行音效處理過程中仍需記憶體63來存取， 使得DSP64所要求之記憶體63容量會大於經降頻單元62 儲存記憶體63中的語音訊號所要求之記憶體63容量，因 此音效處理裝置6所需之記憶體63容量依需較大容量的 DSP64來決定。舉例來說，諸如回音之音效處理所需之記憶 體各量係依要求最大延遲（delay)時間、取樣頻率與各取樣 所需的儲存位元（bit)所決定（記憶體容量=最大延遲時間* 取樣頻率*各取樣之儲存位元），假定最大延遲時間為第9頁 本紙張尺度適用中國國家標準（CNS)A4規格⑵Q χ 297公爱） .. --線. 532045 五、發明說明（ 300ms、取樣頻率為二萬四赫茲以及各取樣之儲存位元為μ 位兀，則圯憶體63容量僅需144〇〇位元組（即 〇·⑽4_峰1152，㈣侧bytes)。由於不管是何種音效 處理決定所需記憶體容量的一重要因素為取樣頻率，如此 相較於習用音效處理裝置，本實施例之音效處理裝置6在 音效處理前先行利用降頻單元62將語音訊號之取樣頻率由 第一頻率（如四萬八赫茲）降低至第二頻率（如二萬四赫 茲），使得音效處理所需之記憶體63容量可隨之大幅降低， 10 15 20 以達到降低所需記憶體63容量之功效，而且隨著語音訊號 之取樣頻率降低至第二頻率，則語音訊號之資料量亦隨之 減少，讓DSP64相較於以往所需要的運算能力要求（指Mlps) 也可達到大幅降低之功效，如此記憶體63與Dsp64所佔用 的電路面積與成本都可大幅降低，以達到更符合電子零件 微小化之趨勢與更容易整合於單一積體電路或單一晶片 上。 此昇頻單元65係電性連接DSP64與混音器66。該昇 頻單兀65用以接收經dSP64音效處理之語音訊號並調昇此 5吾音訊號之取樣頻率以使語音訊號之取樣頻率由第二頻率 恢復至第一頻率，以形成一音效訊號來與原始語音訊號整 合。 此混音器66係電性連接至昇頻單元65與ADC61並用 以接收自ADC61輸出之原始語音訊號（即指未經降頻單元 62、DSP64、昇頻單元65處理之語音訊號）以及接收自昇頻 單凡65輸出之音效訊號（即指經降頻單元62、DSP64、昇頻 第10頁 本氏張尺&舶中國國家標準（CNS)A〗規格（训7297公爱） 532045 15 20 A7 B7 五、發明說明（8 ) 單兀65處理之語音訊號），以將兩訊號合成一聲音訊號輸 出至DAC67。由於未經降頻之原始語音訊號，在本實施例 中，於類比至數位轉換過程中以與習知相同之第一頻率作 為取樣頻率，因其取樣頻率如此高，使得原始語音訊號之 5音貝可視為原音重現般，以使混合此原始語音訊號之聲音 訊號可達到高音質之功效。本實施例中，此混音器66係一 累加器。 ^ 此MC67係電性連接該混音器66並用以將該聲音訊號 由數位Λ號轉換成類比訊號，以適於輸出至該揚聲器5播 10 放。 依刖所述，為了讓本發明更容易被明暸，一併參照第 三圖對本實施例之方法在下文中作說明。 。、，首，步驟71，由ADC61接收由麥克風4擷取之語音訊 號亚以第一頻率（如4萬八赫茲）作為取樣頻率將語音訊號 數位化（即將類比訊號轉為數位訊號）成原始語音訊號，而 後ADC61並將原始語音減分別傳送至降頻單元62與混音 器66。 其後，步驟72,由降頻單元62調降接收的原始語音 訊號的取樣頻率至低於第一頻率之第二頻率。 而後步驟73’降頻單元62更將調降後的語音訊號傳 送至記憶體63中暫存。 緊接著步驟74,由DSP64自記憶體“中存取語音訊 號以依需求進行音效處理，在本實施例中由於音效之語音 訊號的取樣頻率（即第二頻率）大幅降低，使得對記憶體⑽ 【 ，_— 第u頁V. Description of the Invention ([Field of the Invention] The present invention relates to a method and device for processing effects such as ^, in particular, a method for reducing sound effects processing by adjusting the sampling frequency (^ Beifeng (Zhanshi 11 rate)). The sound capacity processing method and device for ordinary microphones are used. 5 [Learning Skills] I microphone sound processing can be roughly divided into two types of sound effects processing methods of analog and digital signals. Compared with analog sound effects at present Processing method, digital sound processing method can get better sound effects. Because microphone 1 is used to convert analog signals into sound waves and speaker 2 is used to convert analog signals into sound waves, the digital sound processing device 3 needs to convert After the signal of the microphone! Is digitized, the sound effect is processed, and after the sound effect is processed, the signal must be analogized to be suitable for the output of the speaker 2. Therefore, as shown in the first figure, it is a digital sound processing device 3 for a conventional microphone. Block diagram. This device 3 includes an analog-to-digital converter that receives an analog signal from the microphone i and digitizes it. (analog-to-digital converter, ADC for short) 31, a memory 32 for temporarily storing digitized voice signals, and a digital signal processor (hereinafter referred to as digital signal processor) for capturing voice signals from the memory 32 for signal processing DSP) 33, and a digital-to-analog converter (DAC) 34 which receives a signal processed by the DSP 33 and analogizes it to be suitable for transmission to the speaker 2. Thus, when the microphone 1 converts sound waves When converting to analog signals, the signal can be digitized and stored in the memory 32 by the ADC31, and then the DSP33 can read the signal from the memory 32 and process it as needed to create various effects, such as echo 10 15 20 page 4 -------- ^ --------- (Please read the notes on the back before filling out this page)% Home Standard (CNSM4 Specification (210x 297 mm) 532045 A7 Intellectual Property Bureau, Ministry of Economic Affairs Printed by Consumer Cooperative Ltd. 5. Description of Invention (2 (eCh〇), Chorus, Flanger, Pitch Shlft, Equalizer, Voice Changer, Compression and Decompression, Voice = And so on, and then by DAC34 The signal is converted into an analog signal and played by: Speaker 2. /, ,, In recent years, with the rise of multimedia entertainment, the 9 qualities required by users for microphones have also increased, which affects one of the sound quality after sound processing. The main factor is the sampling frequency. This sampling frequency refers to the technique of fetching continuous data at a fixed periodic frequency during the conversion of analog to digital signals to 31 to convert analog signals into digital signals. Generally speaking, the sampling frequency needs to be higher than the frequency of the sound source (for a microphone, human voice), and the better the sound quality with the sampling frequency. Today, electronic components tend to be integrated and miniaturized, so that the conventional audio processing device 3 will also be formed as a system integrated in a single integrated circuit, or even integrated on a single chip. However, with the increase of the sampling frequency, the capacity of the memory 32 in the sound processing device 3 and the amount of MIPS (milllon instructi〇ns command se_d; million instructions per second) required by the corresponding Dsp also increase significantly, for example For example, the audio processing rate such as the echo and the storage bit (blt) required for each sample depend on the memory capacity = max = delay time * sampling frequency * storage bit for each sample), assuming a maximum delay time of 300ms , The sampling frequency is 48,000 hertz (Hz) and the storage bit of each sample is 16 bits, then the memory 32 capacity requires 28_ bytes (ie 0.3 * · continued 6 = 23 coffee blts = 2 coffee bytes). In this way, the sound processing device 1 requires such a high capacity of the memory 32 and the operation capacity of the deletion 3 5 10 orders 15 lines 20 V. Description of the invention (3) " " --- so it needs to occupy too much circuit area As a result, it is difficult to integrate the conventional audio processing 3, and even becomes a significant burden on the chip. 2. If the sound of microphone 1 is transmitted to speaker 2 for playback without any sound effect processing, the sampling frequency should be as high as possible, and then played with better sound quality. However, in fact, the faster the sampling speed, the less humans can detect the difference between the two samples'. And because the sound will be changed by sound processing, for example, the echo is a simulated sound bounced through a wall in a large space. The reverberation of the return, such an echo is attenuated by the absorption of the wall and the air, and most of the sound quality is not good, so people do not want high-quality sound directly outputted by sound. For microphone i applications, a sampling frequency of approximately eight kilohertz is more than enough for human sound. Therefore, without affecting the sound quality of the microphone output, if the sampling frequency of the sound processing is reduced, the requirements of the memory 32 capacity and the DSP33 computing capacity can be reduced accordingly, thereby solving the existing problems. 15 [Summary of the Invention] Therefore, an object of the present invention is to provide a sound effect processing method for a microphone that reduces the sampling frequency during the sound effect processing to save the memory capacity and reduce the requirement for computing power. % * Another purpose of the invention is to provide a sound effect processing method for microphones. It is used to reduce the sampling frequency during sound effect processing and restore the original sampling frequency after sound effect processing. Mixing to achieve the effect of reducing the memory capacity required for low-frequency processing while maintaining high sound quality. Another object of the present invention is to provide a sound effect processing device for a microphone that can achieve a reduction in sound effects. 532045 A7. V. Description of the invention (4) Memory capacity required for processing. The purpose of this [month t] is to provide a sound processing device for a microphone that can reduce the requirements for computing the power of the moon. Therefore, the sound effect processing method of the microphone of the present invention is suitable for sound effect processing, and the voice signal extracted by the microphone is transmitted to a speaker for playback. The method includes the following steps... —) The voice signal from the microphone is regarded as one. The sampling frequency of the first frequency is digitally converted into an original voice signal and output;) the sampling frequency of the 5 Hai original voice signal signal in v step A) is lowered for audio processing, and the processed signal is then processed The sampling frequency is restored to the first frequency to form a sound effect signal; and C) the original speech signal of step a) and the sound effect signal of step β) are mixed and mixed into a sound signal, and then the sound is further The signal is analogized to be suitable for output to the speaker for conversion into sound waves for playback. Furthermore, the audio processing device of the present invention includes an analog-to-digital converter, a sound-effect processing unit, a mixer, and a digital-to-analog converter. Among them, the analog-to-digital converter receives the voice signal of the microphone. The first frequency is used as the sampling frequency to digitize the voice signal to form an initial voice signal. The sound effect processing unit receives the original audio signal from the analog-to-digital converter and reduces the sampling frequency of the original voice signal. The sound effect processing and the sampling frequency of the voice signal are restored during the sound effect processing to form a sound effect flood number; the mixer is respectively converted from the analog to digital conversion w and the sound effect processing unit receives the original sound signal and the sound effect signal into one The audio signal; and the digital-to-analog converter is received from the mixer, please order 5 10, order 15 and connect the original connector 20. Page 7 This paper applies the Chinese National Standard (CNS) A4 specification (210 X 297 meals) 532045 15 20 Λ7 V. Description of the invention (5) The sound signal is analogized to the sound signal suitable for rotation to the speaker broadcast Description] Other features and advantages of the present invention will be clearly understood in the following detailed description of the preferred embodiment with reference to the drawings. The first picture is the digital sound processing for a conventional microphone. Circuit block diagram of the device. This device is suitable for receiving and processing the microphone signal and suitable for output to a speaker for playback. The second diagram is a circuit block diagram of a preferred embodiment of the present invention. This implementation is a combination of 10 cases. A microphone and a speaker; and the second diagram is a flowchart of the preferred embodiment of the second diagram. [Detailed description of the preferred embodiment] Referring to the second figure, a preferred embodiment of a sound effect processing device 6 for a microphone of the present invention is shown. The audio processing device 6 is suitable for receiving an analog sound signal captured by a microphone 4 and is suitable for transmitting to a speaker for playback after being processed. The audio processing device 6 of this embodiment includes an analog-to-digital converter (ADC) 6 frequency reduction unit 62, a memory 63, a digital signal processing unit tc (DSP) 64, an up-conversion unit 65, and a mixer. A microphone 66 and a digital-to-analog converter 67. The ADC61 is electrically connected to the microphone 4, the down-converting unit 62, and the mixer 66. In this embodiment, the microphone 4 converts the voice of the sound wave into an analog voice signal, and the ADC 61 converts the analog voice signal output by the microphone 4 into a digital original voice signal at a sampling frequency of a first frequency (for example, 48 kHz). To the frequency reduction unit 62 and the mixer 66. Page 8 This paper scales the title ^ event f 532045 Λ7 B7 V. Description of the invention (6 10 This frequency reduction unit 62 is used to reduce the original voice signal from ΔηΓβ1 ^ gated from ADC61. The second frequency of the first frequency is output to the memory 63. ί In the example, since the sound effect processing m is for the application of the microphone 4, the brother-sampling frequency is only sufficient to cope with human voice, in other words, the first-frequency only needs to be at least It can be equal to eight kilohertz. For example, the second frequency is: 24 hertz or twelve hertz, etc., and the value of the second frequency is determined according to the designer's needs. This memory 63 is used to temporarily store the The frequency unit 62 down-converts the voice signal to the second frequency for subsequent DSP64 to retrieve the voice signal from the memory 63 for proper processing and to allow the DSP64 to access the data during the processing of the voice signal., ^ — (Please (Please read the notes on the back before filling this page) 5 ο 1 2_i_ The Intellectual Property Bureau of the Ministry of Economic Affairs's Consumer Cooperatives printed this DSP64 series to electrically connect the memory 63 and the upscaling unit 64 and have a voice signal processing function to Body 63 Capture digital voice signals to process the phone number data according to the desired effect, and the sound effects processing includes echo, & sound 9, equalizer, equalizer, voice change, compression and decompression, speech recognition and so on. The memory required for voice signal processing will be accessed by the memory 63 during the sound processing by the DSP64, so that the capacity of the memory 63 required by the DSP64 will be greater than that of the voice signal stored in the memory 63 by the frequency reduction unit 62 The required memory 63 capacity, so the memory 63 capacity required by the sound processing device 6 is determined by a larger capacity of the DSP64. For example, the amount of memory required for sound effects processing such as echo is the maximum delay required (Delay) Time, sampling frequency and storage bits required for each sample (memory capacity = maximum delay time * sampling frequency * storage bits for each sample), assuming the maximum delay time is on page 9 Paper size applies Chinese National Standard (CNS) A4 specification ⑵Q χ 297 public love) ..-line. 532045 V. Description of the invention (300ms, sampling frequency is 24,000 Hz and storage of each sample The bit is μ, and the memory 63 requires only 14400 bytes (that is, 0 · ⑽4_peak 1152, side bytes). No matter what kind of sound processing is used to determine the required memory capacity, The important factor is the sampling frequency. Compared with the conventional audio processing device, the audio processing device 6 of this embodiment uses the frequency reduction unit 62 to reduce the sampling frequency of the voice signal from the first frequency (such as 48,000 Hz) before the audio processing. Reduced to the second frequency (such as 24,000 Hz), so that the memory 63 capacity required for audio processing can be greatly reduced, 10 15 20 to achieve the effect of reducing the required memory 63 capacity, and as the voice signal When the sampling frequency is reduced to the second frequency, the data volume of the voice signal is also reduced, so that the DSP64 can achieve a significantly reduced efficiency compared to the required computing power requirements (referring to Mlps). Thus, the memory 63 and Dsp64 The occupied circuit area and cost can be greatly reduced, in order to achieve more in line with the trend of miniaturization of electronic components and easier to integrate on a single integrated circuit or a single chip. The upscaling unit 65 is electrically connected to the DSP 64 and the mixer 66. The upsampling unit 65 is used to receive the dSP64 audio-processed voice signal and raise the sampling frequency of the audio signal to restore the sampling frequency of the voice signal from the second frequency to the first frequency to form an audio signal. Integrated with the original voice signal. This mixer 66 is electrically connected to the up-converting unit 65 and ADC61 and is used to receive the original voice signal output from the ADC61 (that is, the voice signal not processed by the down-converting unit 62, DSP64, and up-converting unit 65) Audio signal output by up-converting single 65 (referring to down-converting unit 62, DSP64, up-converting page 10 Ben's Zhang ruler & China National Standard (CNS) A) specifications (training 7297 public love) 532045 15 20 A7 B7 V. Description of the invention (8) Voice signal processed by unit 65) to synthesize two signals into one audio signal and output to DAC67. Since the original audio signal is not down-converted, in this embodiment, the same first frequency is used as the sampling frequency in the analog-to-digital conversion process. Because the sampling frequency is so high, the 5-tone of the original speech signal Bay can be regarded as the original sound reproduction, so that the sound signal mixed with the original voice signal can achieve the effect of high sound quality. In this embodiment, the mixer 66 is an accumulator. ^ The MC67 is electrically connected to the mixer 66 and is used to convert the sound signal from a digital Λ to an analog signal, suitable for output to the speaker 5 broadcast 10 amplifier. As mentioned, in order to make the present invention easier to understand, the method of this embodiment is described below with reference to the third figure. . First, in step 71, the voice signal captured by the microphone 4 is received by the ADC 61. The first frequency (such as 48 Hz) is used as the sampling frequency to digitize the voice signal (ie, convert the analog signal into a digital signal) into the original voice. Signal, and then the ADC 61 transmits the original speech subtraction to the frequency reduction unit 62 and the mixer 66 respectively. Thereafter, in step 72, the down-sampling unit 62 reduces the sampling frequency of the received original voice signal to a second frequency lower than the first frequency. In step 73 ', the frequency reduction unit 62 further transmits the reduced voice signal to the memory 63 for temporary storage. Immediately following step 74, the DSP64 accesses the voice signal from the memory "to perform sound effect processing as required. In this embodiment, because the sampling frequency (ie, the second frequency) of the voice signal of the sound effect is greatly reduced, the memory is saved. [, _— page u
^本纸張尺度用中國國家標準規格（2】〇 X ---------------------訂·-------I (請先閱讀背面之注意事項再填寫本頁) 532045 五、發明說明（9 需求容量與對DSP64要求運算能力亦隨之大幅降低。 ^後’步驟75調昇經DSP64音效處理之語音訊號的取 ，頻率，使將語音訊號之取樣頻率由第二頻率恢 一 頻率以形成音效訊號。 5 10 15 而後’步驟76，由於經降頻單元62降頻來進行音效 處之訊號在步驟75中調高為原本取樣頻率（即第一頻率）， 使得音效訊號與原始語音訊號皆為同一取樣頻率，如此兩 訊號經混音器66進行混音以形成聲音訊號，而此聲音訊號 由於以高頻（即諸如四萬八赫兹）的第—頻率作為取樣頻率 之原始語音訊號存在’使得聲音訊號之音質可達到至少與 習知音效處理裝置相同之高音質之功效。而且，在本實施 例中音效處理裝置6在決定ADC61之取樣頻率（即第一頻 率），由於較不需顧及後續的音效處理中高取樣頻率而造成 記憶體63容量與DSP64運算能力亦需大幅提升的問題，使 得ADC61之取樣頻率可依設計者需求盡量地提高（例如9萬 六赫兹）’如此使得本發明甚至可達到音質高於習知之功 效。 最後，在步驟77中，由DAC67自混音器66接收聲音 經濟部智慈財產局員工消費合作社印«. 訊號並將此聲音訊號類比化，以適於傳送至揚聲器5播放 20 用。 ° 綜前所述，本發明利用高取樣頻率（第一頻率）來數位 化語音訊號成原始語音訊號，並僅調降欲經音效處理之語 音訊號的取樣頻率和在音效處理後恢復經音效處理之語音 訊號之取樣頻率以與原始語音訊號混音，如此可達到兼顧 第12頁^ This paper uses Chinese national standard specifications (2) 〇X --------------------- Order · --------- I (Please read first Note on the back, please fill out this page again) 532045 V. Description of the invention (9 The required capacity and the computing power required for DSP64 will also be greatly reduced. ^ After step 75, raise the voice signal processing and frequency of DSP64 audio processing, so that The sampling frequency of the voice signal is restored to a frequency from the second frequency to form a sound effect signal. 5 10 15 Then, in step 76, the signal at the sound effect is down-converted by the down-converting unit 62 to increase the original sampling frequency in step 75. (Ie, the first frequency), so that the audio signal and the original voice signal are both at the same sampling frequency, so that the two signals are mixed by the mixer 66 to form a sound signal, and the sound signal is a high frequency (ie, such as 48,000) Hertz) The first frequency as the sampling frequency of the original voice signal exists, so that the sound quality of the sound signal can achieve at least the same high sound quality as the conventional sound effect processing device. Moreover, in this embodiment, the sound effect processing device 6 determines the ADC 61 Sampling frequency (ie first Rate), due to less need to take into account the high sampling frequency in subsequent audio processing, resulting in memory 63 capacity and DSP64 computing power needs to be greatly improved, so that the sampling frequency of ADC61 can be increased as much as the designer needs (such as 96,000 Hertz) 'so that the present invention can even achieve a sound quality higher than the conventional effect. Finally, in step 77, the DAC67 self-mixer 66 receives the sound printed by the staff consumer cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs «. The signal is analogized to be suitable for transmission to the speaker 5 for playback. 20. In summary, the present invention uses a high sampling frequency (the first frequency) to digitize the voice signal into the original voice signal, and only adjusts it for audio processing. The sampling frequency of the voice signal and the sampling frequency of the restored audio signal after audio processing are mixed with the original voice signal.
憶體63容量和DSP64運算能力之功效，進而 降低音效處理裝置6之成本，以使產品 向 惟以上所述者，僅為本發明之較佳實:力。 能以此限定本發明實施之範圍，即大凡， > 田不 範圍及說明書内容所作之簡單的等效變月申：專利 屬本發明專利涵蓋之範圍内。 4飾，皆應仍 第13頁 532045The memory 63's capacity and DSP64's computing power reduce the cost of the sound processing device 6 so that the products described above are only the best practice of the present invention. This can limit the scope of implementation of the present invention, that is, Dafan, > Tian Bu's scope and the simple equivalent of the contents of the description. Patent: The patent belongs to the scope of the invention patent. 4 decorations, all should still page 13 532045
A7 _B7_五、發明說明（11 ) 【元件標號對照】 4麥克風 5揚聲器 6音效處理裝置 61類比至數位轉換器/ ADC 6 2降頻單元 6 3記憶體 64數位訊號處理器/DSP 6 5昇頻單元 6 6混音器 67數位至類比轉換器/DAC ,4 <請先閱讀背面之注意事項再填寫本頁)A7 _B7_ V. Description of the invention (11) [Comparison of component numbers] 4 Microphone 5 Speaker 6 Sound processing device 61 Analog to digital converter / ADC 6 2 Frequency reduction unit 6 3 Memory 64 Digital signal processor / DSP 6 5 liters Frequency unit 6 6 mixer 67 digital to analog converter / DAC, 4 < Please read the notes on the back before filling this page)
裝--------訂---------線IInstall -------- Order --------- Line I
經濟部智慧財產局員工消費合作社印S 第14頁 本紙張尺度適用中國國家標準（CNS)A4規格（210 X 297公釐）Printed by the Consumers' Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs, page 14 This paper is sized for the Chinese National Standard (CNS) A4 (210 X 297 mm)