TWI624709B - Eye glasses with microphone array and method of reducing noise thereof - Google Patents
Eye glasses with microphone array and method of reducing noise thereof Download PDFInfo
- Publication number
- TWI624709B TWI624709B TW103108574A TW103108574A TWI624709B TW I624709 B TWI624709 B TW I624709B TW 103108574 A TW103108574 A TW 103108574A TW 103108574 A TW103108574 A TW 103108574A TW I624709 B TWI624709 B TW I624709B
- Authority
- TW
- Taiwan
- Prior art keywords
- microphone
- channel
- coupled
- audio
- mems
- Prior art date
Links
Landscapes
- Circuit For Audible Band Transducer (AREA)
Abstract
本發明提供一種眼鏡裝置及其對應方法，該眼鏡裝置包含一雜訊消除麥克風陣列。該眼鏡裝置及其使用方法有助於藉由減降會破壞所希望語音的不必要雜訊而利用語音辨識進行人機互動並且因而減少語音中的錯誤。該眼鏡裝置包含：一眼鏡鏡框；一麥克風陣列，其被耦合至該眼鏡鏡框，該麥克風陣列包含至少一第一麥克風與一第二麥克風，該第一麥克風在太陽穴區域附近被耦合至該眼鏡鏡框，該太陽穴區域能夠位在近似於一鏡片開口的頂端角邊與一支撐臂之間，並且該第一麥克風提供一第一音頻通道輸出，而該第二麥克風在該鏡片開口的內緣附近被耦合至該眼鏡鏡框並且提供一第二音頻通道輸出。本發明的實施例進一步包含一第三全向式麥克風與第四全向式麥克風，用以分別提供一第三音頻通道與第四音頻通道。 The invention provides an eyeglass device and a corresponding method thereof, the eyeglass device comprising a noise canceling microphone array. The eyeglass device and method of use thereof facilitates human interaction with speech recognition by reducing unnecessary noise that would disrupt the desired speech and thereby reducing errors in the speech. The eyeglass device includes: a spectacle frame; a microphone array coupled to the spectacle frame, the microphone array including at least a first microphone and a second microphone coupled to the spectacle frame adjacent the temple region The temple area can be positioned between a top corner edge of a lens opening and a support arm, and the first microphone provides a first audio channel output, and the second microphone is adjacent the inner edge of the lens opening Coupled to the spectacle frame and providing a second audio channel output. Embodiments of the present invention further include a third omnidirectional microphone and a fourth omnidirectional microphone for providing a third audio channel and a fourth audio channel, respectively.
Description
本發明大體上關於一種眼鏡裝置及其使用方法，該眼鏡裝置包含一雜訊裝置，該雜訊裝置包含一雜訊消除麥克風陣列 The present invention generally relates to an eyeglass device and a method of using the same, the eyeglass device comprising a noise device comprising a noise canceling microphone array
經由語音辨識使用者介面進行人機互動越來越普及。為促成此人機互動，精確的語音辨識非常有用。然而，來自使用者的環境的不必要雜訊經常會破壞所希望的語音，並且因而增加語音辨識的錯誤並且增加使用者挫折感。具有被耦合至眼鏡鏡框之利用兩個雙麥克風的雜訊消除麥克風陣列的眼鏡裝置的特殊實施例可能太龐大，並且對要求舒適與風格的消費者設計來說太沒有彈性。 Human-computer interaction through voice recognition user interface is becoming more and more popular. To facilitate this human-computer interaction, accurate speech recognition is very useful. However, unnecessary noise from the user's environment often destroys the desired speech and thus increases speech recognition errors and increases user frustration. A particular embodiment of an eyewear device having a noise canceling microphone array utilizing two dual microphones coupled to a spectacle frame may be too bulky and too inflexible for consumer designs that require comfort and style.
所以，本技術需要一種眼鏡，以便克服或最小化上面引述的問題。 Therefore, the present technology requires a pair of glasses to overcome or minimize the problems cited above.
於本發明的其中一實施例中，該眼鏡裝置包含一麥克風陣列，其被耦合至一前鏡框與至少一側鏡框部件中的至少其中一者，該麥克風陣列包含至少一第一麥克風與第二麥克風。該第一麥克風位在太陽穴區域處，介於由該前鏡框所定義的鏡片開口的頂端角邊與該至少一側鏡框部件之間。該第二麥克風位在該鏡片開口的內緣處。第一音頻通道輸出與第 二音頻通道輸出分別延伸自該些第一麥克風與第二麥克風。 In one embodiment of the invention, the eyewear device includes an array of microphones coupled to at least one of a front frame and at least one frame member, the microphone array including at least a first microphone and a second microphone. The first microphone is located at the temple region between the top corner edge of the lens opening defined by the front frame and the at least one frame member. The second microphone is located at the inner edge of the lens opening. First audio channel output and The two audio channel outputs extend from the first microphone and the second microphone, respectively.
於本發明的另一實施例中，該眼鏡裝置包含一數位訊號處理器(DSP)，其具有一射束形成器，該射束形成器被電子連結至該些第一麥克風與第二麥克風，用以接收至少該些第一音頻通道與第二音頻通道並且輸出一主通道以及一或更多參考通道。一聲頻活動偵測器，其被電子連結至該射束形成器，用以接收該主通道與該些參考通道並且輸出一所希望的聲頻活動通道；一適應性雜訊消除器，其被電子連結至該射束形成器與該聲頻活動偵測器，用以接收該主通道、該些參考通道、以及該所希望的聲頻活動通道並且輸出一適應性雜訊消除通道；以及一雜訊減降器，其被電子連結至該適應性雜訊消除器中的聲頻活動偵測器，用以接收該所希望的聲頻活動通道與適應性雜訊消除通道並且輸出一所希望的語音通道。 In another embodiment of the present invention, the glasses apparatus includes a digital signal processor (DSP) having a beam former, the beam former being electronically coupled to the first microphone and the second microphone. And receiving at least the first audio channel and the second audio channel and outputting one main channel and one or more reference channels. An audio activity detector coupled to the beamformer for receiving the main channel and the reference channels and outputting a desired audio activity channel; an adaptive noise canceller, which is electronically Connecting to the beamformer and the audio activity detector for receiving the main channel, the reference channels, and the desired audio activity channel and outputting an adaptive noise cancellation channel; and a noise reduction The drop device is electronically coupled to the audio activity detector in the adaptive noise canceller for receiving the desired audio active channel and the adaptive noise canceling channel and outputting a desired voice channel.
於本發明的又一實施例中包含一種在具有麥克風陣列的眼鏡裝置中減降雜訊的方法，其包含耦合一麥克風陣列至一眼鏡鏡框的步驟，該麥克風陣列包含至少一第一麥克風與第二麥克風。第一麥克風被排列成用以在太陽穴區域附近耦合至該眼鏡鏡框，該太陽穴區域的位置近似介於一鏡片開口的頂端角邊與支撐臂之間。該第二麥克風被排列成用以在該鏡片開口的內緣附近耦合至該眼鏡鏡框。第一音頻通道輸出與第二音頻通道輸出分別由該些第一麥克風與第二麥克風所提供。 A further embodiment of the invention includes a method of reducing noise in a glasses device having a microphone array, the method comprising the steps of coupling a microphone array to a spectacle frame, the microphone array comprising at least a first microphone and a Two microphones. The first microphone is arranged to couple to the spectacle frame adjacent the temple area, the position of the temple area being approximately between the top corner edge of a lens opening and the support arm. The second microphone is arranged to couple to the spectacle frame adjacent the inner edge of the lens opening. The first audio channel output and the second audio channel output are respectively provided by the first microphone and the second microphone.
於本發明的又一實施例中包含藉由使用本文中所述的眼鏡裝置來減降雜訊，並且進一步包含利用一射束形成器來形成射束的步驟，該射束形成器會接收至少該些第一音頻通道與第二音頻通道並且輸出一主通道以及一或更多參考通道。聲頻活動會在一聲頻活動偵測器處被偵測， 該聲頻活動偵測器接收該主通道與該些參考通道並且輸出所希望的聲頻活動通道。雜訊會在一適應性雜訊消除器處被適應性消除，該適應性雜訊消除器接收該主通道、該些參考通道、以及該所希望的聲頻活動通道並且輸出一適應性雜訊消除通道。雜訊會在一雜訊減降器處被減降，該雜訊減降器接收所希望的聲頻活動通道與適應性雜訊消除通道並且輸出一所希望的語音通道。 In yet another embodiment of the invention, the method of reducing noise by using the eyeglass device described herein, and further comprising the step of forming a beam using a beamformer, the beamformer receiving at least The first audio channel and the second audio channel and output a primary channel and one or more reference channels. Audio activity is detected at an audio activity detector. The audio activity detector receives the main channel and the reference channels and outputs a desired audio activity channel. The noise is adaptively eliminated at an adaptive noise canceller that receives the main channel, the reference channels, and the desired audio activity channel and outputs an adaptive noise cancellation aisle. The noise is reduced at a noise eliminator that receives the desired audio activity channel and the adaptive noise cancellation channel and outputs a desired voice channel.
本發明有許多優點。舉例來說，本發明的音頻裝置憑藉該麥克風陣列最小化不必要的雜訊而改良語音的精確辨識，尤其是在運用一數位訊號處理器主動消除不必要雜訊的實施例中，從而在此語音辨識中減少誤差。進一步言之，本發明以每天使用的方便且舒適的形式整合該麥克風陣列與數位訊號處理器。 The invention has many advantages. For example, the audio device of the present invention improves the accurate recognition of speech by minimizing unnecessary noise by the microphone array, especially in an embodiment in which a digital signal processor actively eliminates unnecessary noise, thereby Reduce errors in speech recognition. Further, the present invention integrates the microphone array and digital signal processor in a convenient and comfortable form for everyday use.
100‧‧‧眼鏡裝置 100‧‧‧ glasses device
102‧‧‧眼鏡 102‧‧‧ glasses
104‧‧‧第一麥克風 104‧‧‧First microphone
106‧‧‧第二麥克風 106‧‧‧second microphone
250‧‧‧眼鏡裝置 250‧‧‧ glasses device
252‧‧‧眼鏡 252‧‧‧ glasses
254‧‧‧第一麥克風 254‧‧‧First microphone
256‧‧‧第二麥克風 256‧‧‧second microphone
258‧‧‧第三麥克風 258‧‧‧ third microphone
300‧‧‧麥克風組裝件 300‧‧‧Microphone assembly
302a‧‧‧橡膠長筒第一半部 302a‧‧‧The first half of the rubber long tube
302b‧‧‧橡膠長筒第二半部 302b‧‧‧The second half of the rubber tube
304‧‧‧麥克風 304‧‧‧Microphone
306‧‧‧電線 306‧‧‧Wire
308‧‧‧擋風屏 308‧‧‧ wind screen
350‧‧‧麥克風組裝件 350‧‧‧Microphone assembly
352‧‧‧橡膠長筒 352‧‧‧Rubber long tube
356‧‧‧電線 356‧‧‧Wire
402‧‧‧使用者 402‧‧‧Users
404a-404e‧‧‧位置 404a-404e‧‧‧Location
500‧‧‧眼鏡 500‧‧‧ glasses
500a‧‧‧眼鏡鏡框 500a‧‧‧ glasses frame
501‧‧‧前鏡框 501‧‧‧ front frame
501a-501b‧‧‧末端 End of 501a-501b‧‧‧
501c‧‧‧頂端角邊 501c‧‧‧ top corner
501d-501e‧‧‧內緣 501d-501e‧‧‧ inner edge
502‧‧‧全向式麥克風 502‧‧‧ Omnidirectional microphone
503a‧‧‧側鏡框 503a‧‧‧ side frame
504‧‧‧全向式麥克風 504‧‧‧ Omnidirectional microphone
505‧‧‧接合點 505‧‧‧ joints
506‧‧‧全向式麥克風 506‧‧‧ Omnidirectional microphone
507a-507b‧‧‧開口 507a-507b‧‧‧ openings
507c-507d‧‧‧頂端角邊 507c-507d‧‧‧ top corner
508‧‧‧全向式麥克風 508‧‧‧ Omnidirectional microphone
509‧‧‧橋接支撐區 509‧‧‧Bridge support area
600‧‧‧眼鏡鏡框 600‧‧‧ glasses frame
602‧‧‧第一麥克風 602‧‧‧First microphone
604‧‧‧第二麥克風 604‧‧‧second microphone
606‧‧‧鏡片開口 606‧‧‧ lens opening
700‧‧‧眼鏡裝置 700‧‧‧ glasses device
702‧‧‧上撓性印刷電路板帶 702‧‧‧Upper flexible printed circuit board strip
704‧‧‧第一麥克風 704‧‧‧First microphone
706‧‧‧第四麥克風 706‧‧‧fourth microphone
708‧‧‧下撓性印刷電路板帶 708‧‧‧Under flexible printed circuit board tape
710‧‧‧第二麥克風 710‧‧‧second microphone
712‧‧‧第三麥克風 712‧‧‧ third microphone
800‧‧‧MEMS麥克風器件 800‧‧‧MEMS microphone device
802‧‧‧MEMS麥克風 802‧‧‧ MEMS microphone
804‧‧‧撓性印刷電路板 804‧‧‧Flexible printed circuit board
806‧‧‧墊片 806‧‧‧shims
808‧‧‧裝置外殼 808‧‧‧ device housing
810‧‧‧出口 810‧‧ Export
900‧‧‧麥克風器件 900‧‧‧Microphone device
902‧‧‧麥克風 902‧‧‧ microphone
904‧‧‧聲音埠 904‧‧‧Sound 埠
906‧‧‧撓性印刷電路板 906‧‧‧Flexible printed circuit board
908‧‧‧墊圈 908‧‧‧Washers
910‧‧‧凹腔 910‧‧‧ cavity
912‧‧‧保護部件 912‧‧‧Protection parts
1000‧‧‧眼鏡 1000‧‧‧ glasses
1002‧‧‧麥克風 1002‧‧‧ microphone
1004‧‧‧麥克風 1004‧‧‧ microphone
1006‧‧‧太陽穴區域 1006‧‧‧The temple area
1008‧‧‧前鏡框 1008‧‧‧ front frame
1100‧‧‧方塊圖 1100‧‧‧block diagram
1101‧‧‧雜訊消除電路 1101‧‧‧ Noise Elimination Circuit
1102‧‧‧射束形成(BF)模組 1102‧‧‧beam forming (BF) module
1104‧‧‧適應性雜訊消除(ANC)模組 1104‧‧‧Adaptive Noise Elimination (ANC) Module
1106‧‧‧單訊號雜訊減降(NR)模組 1106‧‧‧Single Signal Noise Reduction (NR) Module
1108‧‧‧聲頻活動偵測(VAD)模組 1108‧‧‧Voice Activity Detection (VAD) Module
1110‧‧‧較近麥克風訊號 1110‧‧‧Close microphone signal
1112‧‧‧較遠麥克風訊號 1112‧‧‧ far microphone signal
1120‧‧‧主訊號 1120‧‧‧ main signal
1122‧‧‧參考訊號 1122‧‧‧Reference signal
1130‧‧‧主訊號 1130‧‧‧ main signal
1132‧‧‧參考訊號 1132‧‧‧Reference signal
1140‧‧‧聲頻活動偵測(DVAD)訊號 1140‧‧‧Digital Activity Detection (DVAD) Signal
1142‧‧‧經雜訊消除的訊號 1142‧‧‧ Signals cancelled by noise
1144‧‧‧所希望的語音 1144‧‧‧ the desired voice
1200‧‧‧方塊圖 1200‧‧‧block diagram
1202‧‧‧射束形成模組 1202‧‧‧beam forming module
1204‧‧‧頻率響應匹配濾波器 1204‧‧‧ frequency response matched filter
1206‧‧‧低通濾波器 1206‧‧‧Low-pass filter
1208‧‧‧低通濾波器 1208‧‧‧Low-pass filter
1210‧‧‧較近麥克風訊號 1210‧‧‧Close microphone signal
1212‧‧‧較遠麥克風訊號 1212‧‧‧ far microphone signal
1220‧‧‧主訊號 1220‧‧‧ main signal
1222‧‧‧參考訊號 1222‧‧‧Reference signal
1230‧‧‧主訊號 1230‧‧‧ main signal
1232‧‧‧參考訊號 1232‧‧‧Reference signal
1300‧‧‧活動方塊圖 1300‧‧‧ activity block diagram
1302‧‧‧聲頻活動偵測模組 1302‧‧‧Audio Activity Detection Module
1304‧‧‧短時功率模組 1304‧‧‧Short-time power module
1306‧‧‧短時功率模組 1306‧‧‧Short-time power module
1308‧‧‧放大器 1308‧‧‧Amplifier
1310‧‧‧放大器 1310‧‧Amplifier
1312‧‧‧組合器 1312‧‧‧ combiner
1314‧‧‧單通道VAD模組 1314‧‧‧Single channel VAD module
1320‧‧‧主訊號 1320‧‧‧ main signal
1322‧‧‧參考訊號 1322‧‧‧Reference signal
1340‧‧‧DVAD訊號 1340‧‧‧DVAD signal
1400‧‧‧方塊圖 1400‧‧‧block diagram
1401‧‧‧雜訊消除電路 1401‧‧‧ Noise Elimination Circuit
1402‧‧‧射束形成(BF)模組 1402‧‧‧beam forming (BF) module
1404‧‧‧適應性雜訊消除(ANC)模組 1404‧‧‧Adaptive Noise Elimination (ANC) Module
1406‧‧‧單通道雜訊減降(NR)模組 1406‧‧‧ Single Channel Noise Reduction (NR) Module
1408‧‧‧聲頻活動偵測(VAD)模組 1408‧‧‧Voice Activity Detection (VAD) Module
1410‧‧‧較近麥克風訊號 1410‧‧‧Close microphone signal
1412‧‧‧第一較遠麥克風訊號 1412‧‧‧first far microphone signal
1414‧‧‧第二較遠麥克風訊號 1414‧‧‧ second far microphone signal
1420‧‧‧第一參考訊號 1420‧‧‧First reference signal
1422‧‧‧主訊號 1422‧‧‧ main signal
1424‧‧‧第二參考訊號 1424‧‧‧second reference signal
1430‧‧‧主訊號 1430‧‧‧ main signal
1432‧‧‧第一參考訊號 1432‧‧‧First reference signal
1434‧‧‧第二參考訊號 1434‧‧‧second reference signal
1440‧‧‧DVAD訊號 1440‧‧‧DVAD signal
1442‧‧‧經雜訊消除的訊號 1442‧‧‧ Signals cancelled by noise
1444‧‧‧所希望的語音 1444‧‧‧ the desired voice
1502‧‧‧音箱管 1502‧‧‧Speaker tube
1504‧‧‧音箱管尖端 1504‧‧‧ Speaker tube tip
1506‧‧‧麥克風 1506‧‧‧ microphone
1508‧‧‧麥克風 1508‧‧‧ microphone
1510‧‧‧麥克風 1510‧‧‧Microphone
1512‧‧‧增益模組 1512‧‧‧ Gain Module
1514‧‧‧延遲模組 1514‧‧‧Delay module
1516‧‧‧增益模組 1516‧‧‧gain module
1518‧‧‧延遲模組 1518‧‧‧Delay module
1520‧‧‧組合器 1520‧‧‧ combiner
1522‧‧‧組合器 1522‧‧‧ combiner
1524‧‧‧右訊號 1524‧‧‧right signal
1526‧‧‧左訊號 1526‧‧‧ Left signal
1652‧‧‧音箱管 1652‧‧‧Speaker tube
1654‧‧‧音箱管尖端 1654‧‧‧ Speaker tube tip
1656‧‧‧麥克風 1656‧‧‧Microphone
1658‧‧‧麥克風 1658‧‧‧Microphone
1660‧‧‧麥克風 1660‧‧‧Microphone
1662‧‧‧麥克風 1662‧‧‧Microphone
1672‧‧‧增益模組 1672‧‧‧ Gain Module
1674‧‧‧延遲模組 1674‧‧‧Delay module
1676‧‧‧增益模組 1676‧‧‧ Gain Module
1678‧‧‧延遲模組 1678‧‧‧Delay module
1680‧‧‧組合器 1680‧‧‧ combiner
1682‧‧‧組合器 1682‧‧‧ combiner
1684‧‧‧右訊號 1684‧‧‧Right signal
1686‧‧‧左訊號 1686‧‧‧ Left signal
1700‧‧‧方塊圖 1700‧‧‧block diagram
1702‧‧‧射束形成模組 1702‧‧‧beam forming module
1704‧‧‧頻率響應匹配濾波器 1704‧‧‧Frequency response matched filter
1706‧‧‧頻率響應匹配濾波器 1706‧‧‧Frequency response matched filter
1710‧‧‧較近麥克風訊號 1710‧‧‧Close microphone signal
1712‧‧‧第一較遠麥克風訊號 1712‧‧‧First far microphone signal
1714‧‧‧第二較近麥克風訊號 1714‧‧‧Second closer microphone signal
1716‧‧‧低通濾波器 1716‧‧‧ low pass filter
1717‧‧‧低通濾波器 1717‧‧‧Low-pass filter
1718‧‧‧低通濾波器 1718‧‧‧ low pass filter
1720‧‧‧主訊號 1720‧‧‧ main signal
1722‧‧‧第一參考訊號 1722‧‧‧First reference signal
1724‧‧‧第二參考訊號 1724‧‧‧second reference signal
1730‧‧‧主訊號 1730‧‧‧ main signal
1732‧‧‧第一參考訊號 1732‧‧‧First reference signal
1734‧‧‧第二參考訊號 1734‧‧‧second reference signal
1800‧‧‧活動方塊圖 1800‧‧‧ activity block diagram
1802‧‧‧聲頻活動偵測(VAD)模組 1802‧‧‧Voice Activity Detection (VAD) Module
1804‧‧‧短時功率模組 1804‧‧‧Short-time power module
1805‧‧‧短時功率模組 1805‧‧‧Short-time power module
1806‧‧‧短時功率模組 1806‧‧‧Short-time power module
1808‧‧‧放大器 1808‧‧‧Amplifier
1809‧‧‧放大器 1809‧‧‧Amplifier
1810‧‧‧放大器 1810‧‧Amplifier
1811‧‧‧組合器模組 1811‧‧‧ combiner module
1812‧‧‧組合器模組 1812‧‧‧ combiner module
1814‧‧‧單通道VAD模組 1814‧‧‧Single channel VAD module
1816‧‧‧單通道VAD模組 1816‧‧‧Single channel VAD module
1818‧‧‧邏輯或閘 1818‧‧‧Logic or gate
1820‧‧‧主訊號 1820‧‧‧ main signal
1822‧‧‧第一參考訊號 1822‧‧‧First reference signal
1824‧‧‧第二參考訊號 1824‧‧‧second reference signal
1840‧‧‧DVAD訊號 1840‧‧‧DVAD signal
圖1所示的係本發明的眼鏡的一實施例的示意圖，其具有兩個嵌入式麥克風。 1 is a schematic illustration of an embodiment of the spectacles of the present invention having two embedded microphones.
圖2所示的係本發明的眼鏡的一實施例的示意圖，其具有三個嵌入式麥克風。 2 is a schematic illustration of an embodiment of the spectacles of the present invention having three embedded microphones.
圖3A所示的係一橡膠長筒的範例***圖。 Figure 3A shows an exemplary exploded view of a rubber long barrel.
圖3B所示的係組裝之後的圖3A的橡膠長筒的示意圖。 Figure 3B is a schematic view of the rubber cartridge of Figure 3A after assembly.
圖4所示的係本發明的一實施例，圖中顯示根據本發明的麥克風的各種選擇性擺放位置。 4 is an embodiment of the present invention showing various selective placement positions of a microphone in accordance with the present invention.
圖5所示的係本發明的另一實施例，其在四個聲音埠處運用四個全向 式麥克風取代兩個雙向式麥克風。 Figure 5 is another embodiment of the present invention, which uses four omnidirectional directions in four sounds The microphone replaces the two two-way microphones.
圖6所示的係本發明的眼鏡的代表圖，其在由眼鏡的前鏡框所定義的鏡片開口中對角線放置兩個全向式麥克風。 Figure 6 is a representative view of the spectacles of the present invention with two omnidirectional microphones diagonally placed in the lens opening defined by the front frame of the spectacles.
圖7所示的係本發明的另一實施例，其在眼鏡的頂端部分與底部部分中放置四個全向式麥克風。 Figure 7 shows another embodiment of the invention in which four omnidirectional microphones are placed in the top and bottom portions of the spectacles.
圖8所示的係適用於本發明的直接聲音埠的其中一實施例的代表圖。 Figure 8 is a representation of one of the embodiments of the direct sound cartridge of the present invention.
圖9所示的係適用於本發明的眼鏡中之具有擋風屏的聲音埠的代表圖。 Fig. 9 is a representative view of a sound cymbal having a windshield suitable for use in the spectacles of the present invention.
圖10所示的係本發明的另一實施例，其中，麥克風面朝內被放置在眼鏡的太陽穴部分處以及面朝下被放置在眼鏡的前鏡框的下方中央角邊處。 Figure 10 is another embodiment of the invention wherein the microphone is placed inwardly at the temple portion of the spectacles and placed face down at the lower central corner of the front frame of the spectacles.
圖11所示的係運用在本發明其中一實施例中的雜訊消除電路的範例實施例的方塊圖。 Figure 11 is a block diagram showing an exemplary embodiment of a noise canceling circuit used in one of the embodiments of the present invention.
圖12所示的係能夠運用在本發明其中一實施例的雜訊消除電路中的射束形成模組的範例實施例的方塊圖。 12 is a block diagram of an exemplary embodiment of a beam forming module that can be utilized in a noise canceling circuit in accordance with one embodiment of the present invention.
圖13所示的係適用於本發明的所希望的聲頻活動偵測模型模組的範例實施例的方塊圖。 13 is a block diagram of an exemplary embodiment of a preferred audio activity detection model module suitable for use with the present invention.
圖14所示的係適用於本發明另一實施例中的雜訊消除電路的範例實施例的方塊圖。 14 is a block diagram of an exemplary embodiment of a noise canceling circuit suitable for use in another embodiment of the present invention.
圖15所示的係根據本發明另一實施例用以儲藏三個麥克風的音箱的射束形成的範例實施例。 Figure 15 is a diagram showing an exemplary embodiment of beam formation for a speaker for storing three microphones in accordance with another embodiment of the present invention.
圖16所示的係根據本發明又一實施例用以儲藏四個麥克風的音箱的射束形成的範例實施例。 Figure 16 is a diagram showing an exemplary embodiment of beam formation for a speaker for storing four microphones in accordance with yet another embodiment of the present invention.
圖17所示的係根據本發明再一實施例用以接受三個訊號的射束形成模 組的範例實施例的方塊圖。 Figure 17 is a diagram showing a beam forming mode for receiving three signals according to still another embodiment of the present invention. A block diagram of an exemplary embodiment of a group.
圖18所示的係根據本發明另一實施例的所希望的聲頻活動偵測模組的範例實施例的方塊圖。 18 is a block diagram of an exemplary embodiment of a desired audio activity detection module in accordance with another embodiment of the present invention.
從隨附圖式中所示本發明的範例實施例的下面更明確說明中會明白前面所述，在全部不同的圖式中，相同的元件符號表示相同部件。該些圖式未必依照比例繪製，取而代之的係將重點放在圖解本發明的實施例。 The same elements are denoted by the same elements throughout the various drawings, and the same reference numerals are used in the various embodiments of the invention. The drawings are not necessarily to scale, the
本發明大體上關於一種具有雜訊消除麥克風陣列的眼鏡裝置，該眼鏡裝置視情況包含一數位訊號處理器；本發明還關於該眼鏡裝置的使用方法。 The present invention generally relates to an eyeglass device having a noise canceling microphone array that optionally includes a digital signal processor; the present invention also relates to a method of using the eyeglass device.
圖1所示的係本發明的眼鏡裝置100的其中一實施例的範例。如圖中所示，眼鏡裝置100包含具有嵌入式麥克風的眼鏡102。眼鏡102有兩個麥克風104與106，第一麥克風104被排列在眼鏡102鏡框的中間。第二麥克風106被排列在眼鏡102鏡框的側邊。麥克風104與106能夠為壓力梯度式麥克風元件，雙向或是單向。每一個麥克風104與106皆係一橡膠長筒裡面的麥克風組裝件，如下面參考圖3A與3B的更明確說明。該橡膠長筒在具有聲音導管的麥克風的前側與背側提供一聲音埠。該兩個麥克風104與106以及它們的個別長筒能夠完全相同。該些麥克風104與106能夠被氣密式密封(舉例來說，封閉式密封)。該些聲音導管會被擋風材料填充。該些埠係被織狀纖維層密封。該些下聲音埠與上聲音埠被一防水薄膜密封。該些麥克風能夠被建構在眼鏡鏡框的結構之中。每一個麥克風有頂端 孔洞與底部孔洞，它們為聲音埠。於一實施例中，該兩個麥克風104與106(它們能夠為壓力梯度式麥克風元件)能夠各自以兩個全向式麥克風來取代。 An example of one embodiment of the eyewear device 100 of the present invention is shown in FIG. As shown in the figures, the eyewear device 100 includes eyeglasses 102 with an embedded microphone. The glasses 102 have two microphones 104 and 106, and the first microphones 104 are arranged in the middle of the frame of the glasses 102. The second microphone 106 is arranged on the side of the frame of the glasses 102. The microphones 104 and 106 can be pressure gradient microphone elements, bidirectional or unidirectional. Each of the microphones 104 and 106 is a microphone assembly within a rubber barrel, as will be more clearly explained below with reference to Figures 3A and 3B. The rubber long barrel provides a sound cymbal on the front side and the back side of the microphone having the sound tube. The two microphones 104 and 106 and their individual barrels can be identical. The microphones 104 and 106 can be hermetically sealed (for example, a hermetic seal). The sound conduits are filled with windshield material. The tethers are sealed by a layer of woven fibers. The lower sounds and the upper sounds are sealed by a waterproof film. The microphones can be constructed in the structure of the spectacle frame. Each microphone has a top Holes and bottom holes, they are squeaky. In one embodiment, the two microphones 104 and 106 (which can be pressure gradient microphone elements) can each be replaced with two omnidirectional microphones.
圖2所示的係本發明一實施例的另一範例。如圖2中所示，眼鏡裝置250包含眼鏡252，其具有三個嵌入式麥克風。圖2的眼鏡252雷同於圖1的眼鏡102；但是，取而代之的係，其運用三個麥克風，而非兩個。圖2的眼鏡252有一第一麥克風254被排列在眼鏡252的中間，一第二麥克風256被排列在眼鏡252的左側，以及一第三麥克風258被排列在眼鏡252的右側。該三個麥克風能夠被運用在上述的三麥克風實施例之中。 Figure 2 shows another example of an embodiment of the invention. As shown in Figure 2, the eyewear device 250 includes eyeglasses 252 having three embedded microphones. The glasses 252 of Figure 2 are identical to the glasses 102 of Figure 1; however, instead, the system uses three microphones instead of two. The glasses 252 of FIG. 2 have a first microphone 254 arranged in the middle of the glasses 252, a second microphone 256 arranged on the left side of the glasses 252, and a third microphone 258 arranged on the right side of the glasses 252. The three microphones can be utilized in the three microphone embodiment described above.
圖3A所示的係本發明的一麥克風組裝件300的***圖。如圖中所示，橡膠長筒302a至b被分成橡膠長筒第一半部302a以及橡膠長筒第二半部302b。麥克風304位於該些橡膠長筒半部之間。每一個橡膠長筒302a至b皆有擋風屏308材料的內襯；然而，圖3A僅在橡膠長筒第二半部302b中顯示該擋風屏。於壓力梯度式麥克風的情況中，該空氣導管以及聲音埠與音箱內部之間的開放空間被擋風泡沫材料(例如，位在橡膠長筒上方的擋風套筒)填充。 3A is an exploded view of a microphone assembly 300 of the present invention. As shown in the figure, the rubber long cylinders 302a to b are divided into a rubber long cylinder first half 302a and a rubber long cylinder second half 302b. A microphone 304 is located between the rubber barrel halves. Each of the rubber elongate tubes 302a-b has an inner lining of the material of the windshield 308; however, Fig. 3A shows the windshield only in the rubber elongate second half 302b. In the case of a pressure gradient microphone, the air duct and the open space between the sound cymbal and the interior of the enclosure are filled with a windshield foam material (eg, a windshield sleeve positioned above the rubber barrel).
麥克風304被排列成擺放在該橡膠長筒的兩個半部302a至b之間。麥克風304與橡膠長筒302a至b的尺寸被設計成使得該麥克風304適配於該橡膠長筒的該些半部302a至b內的一凹腔之中。該麥克風耦合一電線306，該電線延伸至該橡膠長筒302a至b外面並且，舉例來說，會被連接至一雜訊消除電路，例如，下面參考圖11至14所述。 The microphones 304 are arranged to be placed between the two halves 302a to b of the rubber elongate barrel. The microphone 304 and the rubber elongate cylinders 302a-b are sized such that the microphone 304 fits into a recess in the halves 302a-b of the rubber elongate barrel. The microphone is coupled to a wire 306 that extends outside of the rubber barrels 302a-b and, for example, is coupled to a noise cancellation circuit, for example, as described below with reference to Figures 11-14.
圖3B所示的係組裝之後的麥克風組裝件350的透視圖。圖3B中所示的橡膠長筒352有兩個接合在一起的圖3A的半部302a至b，其 中，一麥克風(圖中並未顯示)在其內部。一電線356被耦合至存在於該橡膠長筒352的麥克風，俾使得，舉例來說，其會被連接至下面所述的雜訊消除電路。 Figure 3B is a perspective view of the microphone assembly 350 after assembly. The rubber cartridge 352 shown in Figure 3B has two halves 302a-b of Figure 3A joined together, which Medium, a microphone (not shown) is inside it. A wire 356 is coupled to the microphone present in the rubber barrel 352 such that it will be coupled, for example, to the noise cancellation circuit described below.
圖4所示的係本發明的一實施例400，圖中顯示麥克風的各種選擇性擺放位置404a至e。如上述，該些麥克風為壓力梯度式。於一實施例中，麥克風能夠被放置在圖4中所示的任何位置，或是，圖4中所示位置的任何組合。於一雙麥克風系統中，最靠近使用者嘴巴的麥克風稱為MIC1，而比較遠離使用者嘴巴的麥克風稱為MIC2。於一實施例中，MIC1與MIC2兩者能夠同線放置在位置1 404a處。於其它實施例中，該些麥克風能夠被定位在下面位置處：-MIC1被定位在位置1 404a處且MIC2被定位在位置2 404b處；-MIC1被定位在位置1 404a處且MIC2被定位在位置3 404c處；-MIC1被定位在位置1 404a處且MIC2被定位在位置4 404d處；-MIC1被定位在位置4 404d處且MIC2被定位在位置5 404e處；-MIC1與MIC2兩者皆被定位在位置4 404d處。 4 is an embodiment 400 of the present invention showing various selective placement positions 404a through e of the microphone. As mentioned above, the microphones are pressure gradient type. In one embodiment, the microphone can be placed at any of the locations shown in Figure 4, or any combination of the locations shown in Figure 4. In a dual microphone system, the microphone closest to the user's mouth is called the MIC1, and the microphone that is farther away from the user's mouth is called the MIC2. In one embodiment, both MIC1 and MIC2 can be placed in line 1 at location 1 404a. In other embodiments, the microphones can be positioned at a position where -MIC1 is positioned at position 1 404a and MIC2 is positioned at position 2 404b; -MIC1 is positioned at position 1 404a and MIC2 is positioned at Position 3 404c; -MIC1 is located at position 1 404a and MIC2 is positioned at position 4 404d; -MIC1 is positioned at position 4 404d and MIC2 is positioned at position 5 404e; both MIC1 and MIC2 Positioned at position 4 404d.
倘若位置4 404d有一麥克風的話，其會被運用在一垂飾內。 If position 4 404d has a microphone, it will be used in a pendant.
該些麥克風亦能夠被運用在位置404a至e的其它組合處，或是，被運用在圖4中沒有顯示的位置處。 The microphones can also be used at other combinations of locations 404a through e, or at locations not shown in FIG.
每一個壓力梯度式麥克風元件皆能夠被位在每一個聲音埠之位置處的兩個全向式麥克風取代，從而導致總共有四個麥克風。來自此些雙全向式麥克風的訊號能夠由上述的電子式或數位式射束形成電路系統來處理，用以產生一壓力梯度射束圖樣。此壓力梯度射束圖樣會取代該等效的壓力梯度式麥克風。 Each of the pressure gradient microphone elements can be replaced by two omnidirectional microphones positioned at each of the sound cymbals, resulting in a total of four microphones. Signals from such dual omnidirectional microphones can be processed by the electronic or digital beamforming circuitry described above to produce a pressure gradient beam pattern. This pressure gradient beam pattern replaces the equivalent pressure gradient microphone.
於本發明的一實施例中，倘若運用一壓力梯度式麥克風的話，每一個麥克風會在一橡膠長筒裡面，該橡膠長筒延伸具有聲音導管的麥克風的前側與背側的聲音埠。在該橡膠長筒的末端處，該新聲音埠會對齊該管中的開口，該處的空乏空間會被擋風材料填充。倘若運用兩個全向式麥克風取代一個壓力梯度式麥克風的話，那麼，每一個麥克風的聲音埠則對齊該開口。 In an embodiment of the invention, if a pressure gradient microphone is used, each of the microphones will be inside a rubber tube that extends the front and back sides of the microphone with the sound tube. At the end of the rubber barrel, the new sound cymbal will align with the opening in the tube where the depleted space will be filled with the windshield material. If two omnidirectional microphones are used instead of a pressure gradient microphone, then the sound of each microphone is aligned with the opening.
於一實施例中，一長音箱雙麥克風頭戴式器件(headset)會看來好像習知的近距發話音箱麥克風；但是，其係一種具有平行的兩個麥克風的大音箱。該音箱的一末端麥克風被放置在使用者嘴巴的前面。該近距發話長音箱雙麥克風設計的目標係軍事、航空、工業中的重雜訊用途並且具有無比空前的雜訊消除效能。舉例來說，其中一個主麥克風會直接被定位在嘴巴前面。一第二麥克風會被定位在嘴巴的側邊。該兩個麥克風會有完全相同的外殼。該兩個麥克風能夠平行放置並且垂直於該音箱。每一個麥克風皆有前開口與背開口。DSP電路系統能夠位在該兩個麥克風之間的殼體之中。 In one embodiment, a long speaker dual microphone headset will appear to be a conventional close-talk speaker microphone; however, it is a large speaker with two microphones in parallel. The microphone at the end of the speaker is placed in front of the user's mouth. The dual-microphone design of the close-talking speaker is designed for military, aerospace, and industrial heavy-duty applications and has an unprecedented level of noise cancellation. For example, one of the main microphones will be positioned directly in front of the mouth. A second microphone will be positioned on the side of the mouth. The two microphones will have identical housings. The two microphones can be placed in parallel and perpendicular to the speaker. Each microphone has a front opening and a back opening. The DSP circuitry can be located in a housing between the two microphones.
麥克風被儲藏在一橡膠或矽質固持器(舉例來說，橡膠長筒)中，必要時有一空氣導管延伸至該些聲音埠。該殼體保持該麥克風在一氣 密式容器中並且提供衝擊吸收效果。該些麥克風前埠與背埠被一擋風層覆蓋，其係由用以降低風噪的多個織狀纖維層或是擋風泡沫材料製成。該麥克風塑膠殼體上的外送孔洞會被防水薄膜材料或是特殊防水塗層覆蓋。 The microphone is stored in a rubber or tantalum holder (for example, a rubber long barrel), and if necessary, an air duct extends to the sound cymbals. The housing keeps the microphone in a gas It is also provided in a compact container and provides an impact absorption effect. The front and back of the microphones are covered by a windshield layer, which is made of a plurality of woven fiber layers or windshield foam materials for reducing wind noise. The delivery holes in the plastic housing of the microphone are covered by a waterproof film material or a special waterproof coating.
於另一實施例中，一會議用鵝頸狀麥克風能夠提供雜訊消除。在大型的會議廳中，回音對聲音記錄來說會是一項問題。被麥克風記錄的回音會造成咆嘯聲(howling)。嚴重的回音會讓使用者無法調高揚聲器音量並且造成有限的可聽性。會議廳與會議室會在牆上裝飾昂貴的吸音材料用以降低回音以達較高的揚聲器音量並且在全部的聽眾中提供均勻的聲音場分佈。電子回音消除配備能夠被用來降低回音並且提高揚聲器音量；但是，此配備相當昂貴、可能有裝配難度、以及經常需要用到聲學專家。 In another embodiment, a conference gooseneck microphone can provide noise cancellation. In a large conference hall, echoing can be a problem for sound recording. The echo recorded by the microphone can cause howling. A severe echo can prevent the user from turning up the speaker volume and causing limited audibility. Meeting rooms and meeting rooms decorate the wall with expensive sound absorbing materials to reduce echo to achieve higher speaker volume and provide a uniform sound field distribution throughout the audience. Electronic echo cancellation equipment can be used to reduce echo and increase speaker volume; however, this equipment is quite expensive, may be difficult to assemble, and often requires the use of acoustic experts.
於一實施例中，一雙麥克風雜訊消除會議用麥克風能夠提供低廉且容易施行的解決方式來解決會議廳或會議室中的回音問題。上述的雙麥克風系統能夠被放置在一桌上型鵝頸狀麥克風之中。管中的每一個麥克風皆係一壓力梯度式雙向、單向、或是超方向(super-directional)麥克風。 In one embodiment, a dual microphone noise cancellation conference microphone can provide an inexpensive and easy to implement solution to the echo problem in a conference hall or conference room. The dual microphone system described above can be placed in a tabletop gooseneck microphone. Each microphone in the tube is a pressure gradient bidirectional, unidirectional, or super-directional microphone.
圖5所示的係本發明的眼鏡500的一實施例。眼鏡500包含眼鏡鏡框500a，其具有用以定義反向末端501a和501b的前鏡框501，及靠近反向末端501a、501b中至少一者的至少一頂端角邊501c，及內緣501d、501e。內緣501d、501e定義開口507a、507b，其之間跨具有橋接支撐區509。開口507a、507b包含頂端角邊507c、507d。四個全向式麥克風502、504、506、508以及電子式射束操控取代圖1中所示的兩個雙向式麥克風。以四個全向式麥克風取代兩個雙向式麥克風讓眼鏡鏡框設計者有更大的彈性以及製造能力。在具有四個全向式麥克風的範例實施例中，該四個全向式麥 克風能夠位於眼鏡鏡框的任何位置，較佳的係，該些麥克風對會以鏡片為基準垂直對齊。全向式麥克風506位在部件為前鏡框501和側鏡框503a之間的接合點505處。於此實施例中，兩個全向式麥克風502與504位在內緣501d處且為主麥克風，用以偵測要與干擾分離的主聲音。全向式麥克風506與508為參考麥克風，它們係偵測要與主聲音分離的背景雜訊。該麥克風陣列能夠為多個全向式麥克風，其中，該些全向式麥克風能夠為下面的任何組合：駐極式電容麥克風、類比式微機電系統(MEMS)麥克風、或是數位式MEMS麥克風。 An embodiment of the spectacles 500 of the present invention is shown in FIG. The spectacles 500 include a spectacle frame 500a having a front frame 501 defining opposite ends 501a and 501b, and at least one apex 501c adjacent at least one of the inverted ends 501a, 501b, and inner edges 501d, 501e. The inner edges 501d, 501e define openings 507a, 507b with a bridging support region 509 therebetween. The openings 507a, 507b include top end corners 507c, 507d. Four omnidirectional microphones 502, 504, 506, 508 and electronic beam steering replace the two two-way microphones shown in FIG. Replacing two two-way microphones with four omnidirectional microphones gives the frame designer more flexibility and manufacturing power. In an exemplary embodiment with four omnidirectional microphones, the four omnidirectional wheats The wind can be located anywhere on the eyeglass frame, preferably the pair of microphones are vertically aligned with respect to the lens. The omnidirectional microphone 506 is located at a junction 505 between the front frame 501 and the side frame 503a. In this embodiment, two omnidirectional microphones 502 and 504 are located at the inner edge 501d and are the main microphones for detecting the main sound to be separated from the interference. The omnidirectional microphones 506 and 508 are reference microphones that detect background noise to be separated from the main sound. The microphone array can be a plurality of omnidirectional microphones, wherein the omnidirectional microphones can be any combination of the following: an electret condenser microphone, an analog microelectromechanical system (MEMS) microphone, or a digital MEMS microphone.
圖6中所示的本發明的另一範例實施例包含一具有雜訊消除麥克風陣列的眼鏡裝置，該眼鏡裝置包含：一眼鏡鏡框600；一麥克風陣列，其被耦合至該眼鏡鏡框，該麥克風陣列包含至少一第一麥克風602與一第二麥克風604，該第一麥克風在太陽穴區域附近被耦合至該眼鏡鏡框，該太陽穴區域能夠位在近似於一鏡片開口的頂端角邊與一支撐臂之間，並且該第一麥克風提供一第一音頻通道輸出，而該第二麥克風在該鏡片開口的內側下方角邊附近被耦合至該眼鏡鏡框並且提供一第二音頻通道輸出。該第二麥克風雖然位在鏡片開口606的對角線處；不過，其亦能夠被定位在鏡片的內鏡框中的任何位置處，舉例來說，下方角邊、上方角邊、或是內鏡框邊緣。進一步言之，該第二麥克風能夠沿著鏡片的內緣位在鼻橋的左邊或右邊。 Another exemplary embodiment of the present invention shown in FIG. 6 includes an eyeglass device having a noise canceling microphone array, the eyeglass device comprising: a eyeglass frame 600; a microphone array coupled to the eyeglass frame, the microphone The array includes at least a first microphone 602 and a second microphone 604 coupled to the spectacle frame adjacent the temple region, the temple region being positionable at a top corner edge of a lens opening and a support arm And the first microphone provides a first audio channel output, and the second microphone is coupled to the eyeglass frame near the inner lower corner of the lens opening and provides a second audio channel output. The second microphone is located at the diagonal of the lens opening 606; however, it can also be positioned anywhere in the inner frame of the lens, for example, the lower corner, the upper corner, or the inner frame edge. Further, the second microphone can be positioned to the left or right of the nose bridge along the inner edge of the lens.
於本發明的又一實施例中，該麥克風陣列會利用至少一撓性印刷電路板(PCB)帶被耦合至眼鏡鏡框，如圖7中所示。於此實施例中，本發明的眼鏡裝置700包含：上撓性PCB帶702，其包含第一麥克風704與第 四麥克風706；以及下撓性PCB帶708，其包含第二麥克風710與第三麥克風712。 In yet another embodiment of the invention, the microphone array is coupled to the spectacle frame using at least one flexible printed circuit board (PCB) strip, as shown in FIG. In this embodiment, the eyewear device 700 of the present invention includes: an upper flexible PCB tape 702 including a first microphone 704 and a A four microphone 706; and a lower flexible PCB strip 708 comprising a second microphone 710 and a third microphone 712.
於本發明的進一步範例實施例中，該眼鏡鏡框會進一步包含一對應於該麥克風陣列的出口陣列。該麥克風陣列能夠係底部埠微機電系統(MEMS)麥克風或是頂端埠微機電系統(MEMS)麥克風。如在圖8中所見(其係圖7的眼鏡的麥克風器件)，MEMS麥克風器件800包含MEMS麥克風802，其被固定至撓性印刷電路板804。墊片806分離撓性PCB 804與裝置外殼808。出口810係由撓性PCB 804、墊片806、以及裝置外殼808來定義。出口810係一條音頻管道，用以將音波輸送至MEMS麥克風802。第一MEMS麥克風與第四MEMS麥克風會被耦合至上撓性PCB帶，第二MEMS麥克風與第三MEMS麥克風會被耦合至下撓性PCB帶，而且該MEMS麥克風陣列會被排列成使得該些底部埠或頂端埠經由該些對應出口接收聲音訊號。 In a further exemplary embodiment of the invention, the spectacle frame further includes an array of outlets corresponding to the array of microphones. The microphone array can be a bottom 埠 microelectromechanical system (MEMS) microphone or a top 埠 microelectromechanical system (MEMS) microphone. As seen in FIG. 8 (which is the microphone device of the glasses of FIG. 7), MEMS microphone device 800 includes a MEMS microphone 802 that is secured to flexible printed circuit board 804. Shim 806 separates flexible PCB 804 from device housing 808. The outlet 810 is defined by a flexible PCB 804, a shim 806, and a device housing 808. The outlet 810 is an audio conduit for delivering sound waves to the MEMS microphone 802. The first MEMS microphone and the fourth MEMS microphone are coupled to the upper flexible PCB strip, the second MEMS microphone and the third MEMS microphone are coupled to the lower flexible PCB strip, and the MEMS microphone array is arranged such that the bottom The 埠 or top 接收 receives the sound signal via the corresponding outlets.
本發明的另一實施例會進一步包含一薄膜，舉例來說，用以擋風且擋雨。舉例來說，如能夠在圖9中所見，麥克風器件900包含麥克風902，其定義聲音埠904。撓性印刷電路板906被固定至麥克風902。墊圈908疊置在撓性PCB 906上方，並且舉例來說，能夠由矽質橡膠來製造。撓性PCB 906與墊圈908會定義凹腔910。保護部件912疊置在墊圈908上方，用以阻擋環境因子，例如，風和雨。或者，部件912亦能夠在底部埠外殼中被夾設在眼鏡鏡框與PCB帶之間；或是，在頂端埠外殼中被夾設在眼鏡鏡框與MEMS麥克風之間。 Another embodiment of the present invention may further comprise a film, for example, for windshield and rain. For example, as can be seen in FIG. 9, microphone device 900 includes a microphone 902 that defines sound 埠 904. The flexible printed circuit board 906 is fixed to the microphone 902. The gasket 908 is stacked over the flexible PCB 906 and, for example, can be fabricated from a enamel rubber. Flexible PCB 906 and gasket 908 define a cavity 910. A protective component 912 is placed over the gasket 908 to block environmental factors such as wind and rain. Alternatively, the component 912 can also be sandwiched between the spectacle frame and the PCB strip in the bottom crucible housing or between the spectacle frame and the MEMS microphone in the top end bezel.
圖10所示的係眼鏡1000的另一替代實施例，其中，麥克風1002、1004分別被放置在太陽穴區域1006以及前鏡框1008處。 Another alternative embodiment of the spectacles 1000 shown in FIG. 10, wherein the microphones 1002, 1004 are placed at the temple region 1006 and the front frame 1008, respectively.
圖11所示的係本發明中所運用的一雜訊消除電路1100的範例實施例的方塊圖。如圖中所示，來自兩個麥克風的訊號1110與1112被數位化並且被饋送至雜訊消除電路1101。雜訊消除電路1101能夠為一數位訊號處理(DSP)單元(舉例來說，在一處理器上執行的軟體、硬體方塊、或是多個硬體方塊)。於一實施例中，雜訊消除電路1101會係一數位訊號處理(DSP)晶片、一晶片上系統(SOC)、一藍牙晶片、一具有DSP晶片的聲頻編解碼器、…等。雜訊消除電路1101能夠被放置在靠近使用者耳部的藍牙頭戴式器件中、被放置在有電池的同線控制殼中、或是被放置在連接器內、…等。雜訊消除電路1101能夠由該頭戴式器件所連接的裝置的電池(圖中並未顯示)或是電源(圖中並未顯示)供電，例如，該裝置的電池、或是來自USB、micro-USB、或是Lightening連接器的電力。 Figure 11 is a block diagram of an exemplary embodiment of a noise cancellation circuit 1100 utilized in the present invention. As shown in the figure, signals 1110 and 1112 from the two microphones are digitized and fed to the noise cancellation circuit 1101. The noise cancellation circuit 1101 can be a digital signal processing (DSP) unit (for example, a software, a hardware block, or a plurality of hardware blocks executed on a processor). In one embodiment, the noise cancellation circuit 1101 is a digital signal processing (DSP) chip, a system on a chip (SOC), a Bluetooth chip, an audio codec with a DSP chip, and the like. The noise cancellation circuit 1101 can be placed in a Bluetooth headset that is adjacent to the user's ear, placed in the same-line control housing with the battery, or placed in the connector, etc. The noise cancellation circuit 1101 can be powered by a battery (not shown) of the device connected to the head device or a power source (not shown), for example, a battery of the device, or from a USB or a micro. -USB, or the power of the Lightening connector.
雜訊消除電路1101包含四個功能方塊：一射束形成(BF)模組1102、一所希望的聲頻活動偵測(VAD)模組1108、一適應性雜訊消除(ANC)模組1104、以及一單訊號雜訊減降(NR)模組1106。該兩個訊號1110與1112被饋送至BF模組1102，其會產生一主訊號1130與一參考訊號1132給ANC模組1104。較近麥克風訊號1110係收集自較靠近使用者嘴巴的麥克風，而較遠的麥克風訊號則收集自相對遠離使用者嘴巴的麥克風。BF模組1102還會產生一主訊號1120與一參考訊號1122給所希VAD模組1108。於特定的實施例中，主訊號1120與參考訊號1122不同於為ANC模組1104所產生的主訊號1130與參考訊號1132。 The noise cancellation circuit 1101 includes four functional blocks: a beam forming (BF) module 1102, a desired audio activity detection (VAD) module 1108, and an adaptive noise cancellation (ANC) module 1104. And a single signal noise reduction (NR) module 1106. The two signals 1110 and 1112 are fed to the BF module 1102, which generates a main signal 1130 and a reference signal 1132 to the ANC module 1104. The closer microphone signal 1110 is collected from a microphone that is closer to the user's mouth, while the farther microphone signal is collected from a microphone that is relatively far from the user's mouth. The BF module 1102 also generates a main signal 1120 and a reference signal 1122 for the VAD module 1108. In a specific embodiment, the main signal 1120 and the reference signal 1122 are different from the main signal 1130 and the reference signal 1132 generated by the ANC module 1104.
ANC模組1104會處理主訊號1130與參考訊號1132，用以消除該兩個訊號中的雜訊並且輸出一經雜訊消除的訊號1142給單通道NR 模組1106。該單訊號NR模組1106會後置處理來自ANC模組1104的經雜訊消除訊號1142，用以移除任何進一步的殘留雜訊。同時，VAD模組1108會從主訊號1120與參考訊號1122中推知一所希望的聲頻活動偵測(Desired Voice Activity Detection，DVAD)訊號1140，其表示主訊號1120與參考訊號1122中有語音存在或是沒有語音存在。DVAD訊號1140接著會依照BF模組1102的結果被用來控制ANC模組1104與NR模組1106。DVAD訊號1140會向ANC模組1104與單通道NR模組1106表示該訊號中哪一個部分有聲頻資料要分析，其能夠藉由忽略該訊號中沒有聲頻資料的部分而提高ANC模組1104與單通道NR模組1106的處理效率。所希望的語音訊號1144會由單通道NR模組1106來產生。 The ANC module 1104 processes the main signal 1130 and the reference signal 1132 to cancel the noise in the two signals and output a noise-cancelled signal 1142 to the single-channel NR. Module 1106. The single signal NR module 1106 will post-process the noise canceling signal 1142 from the ANC module 1104 to remove any further residual noise. At the same time, the VAD module 1108 infers a desired Desired Voice Activity Detection (DVAD) signal 1140 from the main signal 1120 and the reference signal 1122, which indicates that there is voice in the main signal 1120 and the reference signal 1122 or There is no voice present. The DVAD signal 1140 is then used to control the ANC module 1104 and the NR module 1106 in accordance with the results of the BF module 1102. The DVAD signal 1140 will indicate to the ANC module 1104 and the single channel NR module 1106 which part of the signal has audio data to be analyzed, which can improve the ANC module 1104 and the single by ignoring the portion of the signal that has no audio data. The processing efficiency of the channel NR module 1106. The desired voice signal 1144 is generated by the single channel NR module 1106.
於一實施例中，BF模組1102、ANC模組1104、單NR減降模組1106、以及所希望的VAD模組1108會運用線性處理(舉例來說，線性濾波器)。一線性系統(其運用線性處理)會滿足疊加特性以及縮放特性或齊次性(homogeneity)。疊加特性的意義為該系統的輸出和輸入成正比。舉例來說，倘若滿足下面關係，函數F(x)便為一線性系統：F(x1+x2+…)=F(x1)+F(x2)+… In one embodiment, the BF module 1102, the ANC module 1104, the single NR reduction module 1106, and the desired VAD module 1108 utilize linear processing (eg, a linear filter). A linear system (which uses linear processing) will satisfy the superposition characteristics as well as the scaling characteristics or homogeneity. The meaning of the superposition feature is that the output and input of the system are proportional. For example, if the following relationship is satisfied, the function F(x) is a linear system: F(x 1 +x 2 +...)=F(x 1 )+F(x 2 )+...
倘若輸出和輸入成縮放正比則滿足一度的縮放特性或齊次性。舉例來說，倘若滿足下面關係，函數F(x)便滿足縮放特性或齊次性，其中，α為純量：F(αx)=αF(x) If the output and the input are proportional to the scaling, then the one-degree scaling characteristic or homogeneity is satisfied. For example, if the following relationship is satisfied, the function F(x) satisfies the scaling property or homogeneity, where α is a scalar quantity: F(αx)=αF(x)
相反地，非線性系統則不滿足前面兩項條件。 Conversely, nonlinear systems do not satisfy the first two conditions.
先前的雜訊消除系統運用非線性處理。藉由使用線性處理， 提高輸入會成正比地改變輸出。然而，在非線性處理中，提高輸入則會不正比地改變輸出。使用線性處理會因改良特徵抽出而提供語音辨識的優點。揚聲器辨識演算法係以在安靜的環境中沒有失真被記錄的無雜訊聲頻為基礎所開發。線性雜訊消除演算法不會在經雜訊消除的語音中引進非線性失真。語音辨識能夠應付語音中的線性失真，但卻無法應付語音的非線性失真。線性雜訊消除演算法「通透(transparent)」於語音辨識引擎。在非線性失真雜訊的變異中訓練語音辨識係不可能的行為。非線性失真會中斷用於語音辨識所需要的特徵抽出。 Previous noise cancellation systems used nonlinear processing. By using linear processing, Increasing the input will change the output proportionally. However, in nonlinear processing, increasing the input will change the output in an unbiased manner. The use of linear processing provides the advantage of speech recognition due to improved feature extraction. The speaker recognition algorithm was developed based on noise-free audio that was recorded without distortion in a quiet environment. The linear noise cancellation algorithm does not introduce nonlinear distortion into the noise-cancelled speech. Speech recognition can cope with linear distortion in speech, but it cannot cope with the nonlinear distortion of speech. The linear noise cancellation algorithm is "transparent" to the speech recognition engine. Training speech recognition is an impossible behavior in the variation of nonlinear distortion noise. Nonlinear distortion interrupts the feature extraction required for speech recognition.
線性系統的一種範例為Weiner濾波器，其係一種線性單通道雜訊移除濾波器。Weiner濾波器係一種用於藉由對已觀察到之有雜訊的處理進行線性非時變濾波而產生一所希望的或目標隨機處理之預測值的濾波器，其假設有已知的靜態訊號、雜訊頻譜、以及相加性雜訊。Weiner濾波器會最小化該被預測隨機處理與該所希望處理之間的均方差(mean square error)。 An example of a linear system is the Weiner filter, which is a linear single channel noise removal filter. A Weiner filter is a filter for generating a desired or target random processed prediction value by linearly time-varying filtering of the observed noise processing, assuming a known static signal , noise spectrum, and additive noise. The Weiner filter minimizes the mean square error between the predicted random process and the desired process.
圖12所示的係能夠運用在本發明的雜訊消除電路1101中的射束形成模組1202的範例實施例的方塊圖1200。該BF模組1202接收較近麥克風訊號1210與較遠(也就是，比較遠離所希望的聲音)麥克風訊號1212。 12 is a block diagram 1200 of an exemplary embodiment of a beam forming module 1202 that can be utilized in the noise cancellation circuit 1101 of the present invention. The BF module 1202 receives the microphone signal 1212 that is closer to the microphone signal 1210 and is farther (ie, farther away from the desired sound).
較遠(也就是，比較遠離所希望的聲音)麥克風訊號1212被輸入至一頻率響應匹配濾波器1204。該頻率響應匹配濾波器1204會調整該較遠麥克風訊號1212的頻率響應的增益、相位、以及形狀。舉例來說，該頻率響應匹配濾波器1204能夠針對該兩個麥克風之間的距離來調整該訊號，俾使得代表該較遠麥克風訊號1212的一被輸出的參考訊號1232能夠連同代 表該較近麥克風訊號1210的主訊號1230被處理。主訊號1230與參考訊號1232會被送至ANC模組。 The microphone signal 1212 is input to a frequency response matched filter 1204 farther (i.e., away from the desired sound). The frequency response matched filter 1204 adjusts the gain, phase, and shape of the frequency response of the far microphone signal 1212. For example, the frequency response matching filter 1204 can adjust the signal for the distance between the two microphones, so that an output reference signal 1232 representing the far microphone signal 1212 can be combined with the generation. The main signal 1230 of the closer microphone signal 1210 is processed. The main signal 1230 and the reference signal 1232 are sent to the ANC module.
較近(也就是，比較靠近所希望的聲音)麥克風訊號1210會被輸出至ANC模組，作為主訊號1230。該較近麥克風訊號1210也會被輸入至一低通濾波器1206。該參考訊號1232會被輸入低通濾波器1208，用以創造一被送往所希望的VAD模組的參考訊號1222。於其中一實施例中，舉例來說，低通濾波器1206與1208會調整「近距發話情況」中的訊號，舉例來說，有從2kHz至4kHz的逐漸下降(low off)。不過，在不同的設計以及麥克風與使用者嘴巴相隔不同的距離中亦能夠使用其它頻率。 The microphone signal 1210 is output to the ANC module as the main signal 1230. The nearer microphone signal 1210 is also input to a low pass filter 1206. The reference signal 1232 is input to a low pass filter 1208 for creating a reference signal 1222 that is sent to the desired VAD module. In one embodiment, for example, low pass filters 1206 and 1208 adjust the signal in the "close-range speech situation", for example, from 2 kHz to 4 kHz. However, other frequencies can be used in different designs and at different distances between the microphone and the user's mouth.
圖13所示的係所希望的聲頻活動偵測模組1302的範例實施例的方塊圖。DVAD模組1302會從射束形成模組處接收一主訊號1320與一參考訊號1322。主訊號1320與參考訊號1322會被個別的短時功率模組1304與1306處理。該些短時功率模組1304與1306會包含方均根(RMS)偵測器、功率(PWR)偵測器、或是能量偵測器。該些短時功率模組1304與1306會輸出訊號至個別的放大器1308與1310。該些放大器能夠為對數轉換器(或是對數放大器)。該些對數轉換器1308與1310會輸出至一組合器1312。組合器1312被配置成用以組合多個訊號，例如，組合該主訊號以及該至少一參考訊號中的其中一者，用以藉由以該主訊號的偵測值減去該參考訊號的偵測值(或者，反之亦可)來產生一聲頻活動差值訊號。該聲頻活動差值訊號會被輸入至一單通道VAD模組1314。該單通道VAD模組能夠為一習知的VAD模組。該單通道VAD模組1314會輸出所希望的聲頻活動訊號。 A block diagram of an exemplary embodiment of a desired audio activity detection module 1302 is shown in FIG. The DVAD module 1302 receives a main signal 1320 and a reference signal 1322 from the beam forming module. Main signal 1320 and reference signal 1322 are processed by individual short time power modules 1304 and 1306. The short-term power modules 1304 and 1306 may include a square root mean square (RMS) detector, a power (PWR) detector, or an energy detector. The short time power modules 1304 and 1306 output signals to the respective amplifiers 1308 and 1310. The amplifiers can be logarithmic converters (or logarithmic amplifiers). The logarithmic converters 1308 and 1310 are output to a combiner 1312. The combiner 1312 is configured to combine a plurality of signals, for example, combining one of the main signal and the at least one reference signal for subtracting the reference signal by detecting the detected value of the main signal The measured value (or vice versa) is used to generate an audio activity difference signal. The audio activity difference signal is input to a single channel VAD module 1314. The single channel VAD module can be a conventional VAD module. The single channel VAD module 1314 outputs the desired audio activity signal.
圖14所示的係一雜訊消除電路1401的範例實施例的方塊圖 1400，其被運用來分別接收一較近麥克風訊號1410以及第一較遠麥克風訊號1412與第二較遠麥克風訊號1414。該雜訊消除電路1401雷同於配合圖1所述的雜訊消除電路1101；然而，雜訊消除電路1401被用來接收三個訊號，而非兩個訊號。一射束形成(BF)模組1402被排列成用以接收訊號1410、1412、以及1414，並且輸出一主訊號1430、一第一參考訊號1432、以及一第二參考訊號1434給一適應性雜訊消除模組1404。該射束形成模組會進一步被配置成用以輸出一主訊號1422、第一參考訊號1420、以及第二參考訊號1424給一聲頻活動偵測(VAD)模組1408。 A block diagram of an exemplary embodiment of a noise cancellation circuit 1401 shown in FIG. 1400, which is operative to receive a closer microphone signal 1410 and a first far microphone signal 1412 and a second far microphone signal 1414, respectively. The noise cancellation circuit 1401 is identical to the noise cancellation circuit 1101 described in FIG. 1; however, the noise cancellation circuit 1401 is used to receive three signals instead of two signals. A beam forming (BF) module 1402 is arranged to receive signals 1410, 1412, and 1414, and output a main signal 1430, a first reference signal 1432, and a second reference signal 1434 for an adaptive miscellaneous The signal cancellation module 1404. The beam forming module is further configured to output a main signal 1422, a first reference signal 1420, and a second reference signal 1424 to an audio activity detection (VAD) module 1408.
ANC模組1404會產生一經雜訊消除的訊號1442給單通道雜訊減降(NR)模組1406，其雷同於圖1的ANC模組1104。該單通道NR模組1406接著會輸出所希望的語音1444。VAD模組1408會輸出DVAD訊號給ANC模組1404以及單通道NR模組1406。 The ANC module 1404 generates a noise canceled signal 1442 to the single channel noise reduction (NR) module 1406, which is similar to the ANC module 1104 of FIG. The single channel NR module 1406 then outputs the desired speech 1444. The VAD module 1408 outputs DVAD signals to the ANC module 1404 and the single channel NR module 1406.
圖15所示的係儲藏三個麥克風1506、1508、以及1510的音箱管1502的射束形成的範例實施例。一第一麥克風1506被排列成最靠近音箱管1502的尖端1504，一第二麥克風1508被排列在音箱管1502中比較遠離尖端1504，而一第三麥克風1510被排列在音箱管1502中更遠離尖端1504。第一麥克風1506與第二麥克風1508被排列成用以提供資料以輸出一左訊號1526。該第一麥克風被排列成用以輸出其訊號給一增益模組1512與一延遲模組1514，其會被輸出至一組合器1522。該第二麥克風直接被連接至組合器1522。組合器1522會相減該兩個已提供的訊號，用以消除雜訊，其會創造左訊號1526。 An exemplary embodiment of beam formation for storing three microphones 1506, 1508, and 1510 of speaker tube 1502 is shown in FIG. A first microphone 1506 is arranged closest to the tip end 1504 of the speaker tube 1502, a second microphone 1508 is arranged in the speaker tube 1502 relatively farther from the tip 1504, and a third microphone 1510 is arranged in the speaker tube 1502 further away from the tip. 1504. The first microphone 1506 and the second microphone 1508 are arranged to provide data to output a left signal 1526. The first microphone is arranged to output its signal to a gain module 1512 and a delay module 1514, which are output to a combiner 1522. This second microphone is directly connected to the combiner 1522. The combiner 1522 subtracts the two provided signals to eliminate noise, which creates a left signal 1526.
同樣地，第二麥克風1508被連接至一增益模組1516與一延 遲模組1518，其會被輸出至一組合器1520。第三麥克風1510直接被連接至組合器1520。組合器1520會相減該兩個已提供的訊號，用以消除雜訊，其會創造右訊號1524。 Similarly, the second microphone 1508 is connected to a gain module 1516 and a delay The late module 1518, which will be output to a combiner 1520. The third microphone 1510 is directly connected to the combiner 1520. The combiner 1520 subtracts the two provided signals to eliminate noise, which creates a right signal 1524.
圖16所示的係儲藏四個麥克風1656、1658、1660、以及1662的音箱管1652的射束形成的範例實施例。一第一麥克風1656被排列成最靠近音箱管1652的尖端1654，一第二麥克風1658被排列在音箱管1652中比較遠離尖端1654，一第三麥克風1660被排列在音箱管1652中更遠離尖端1654，以及一第四麥克風1662被排列在音箱管1652中最遠離尖端1654。第一麥克風1656與第二麥克風1658被排列成用以提供資料以輸出一左訊號1686。該第一麥克風被排列成用以輸出其訊號給一增益模組1672與一延遲模組1674，其會被輸出至一組合器1682。該第二麥克風直接被連接至組合器1682。組合器1682會相減該兩個已提供的訊號，用以消除雜訊，其會創造左訊號1686。 An exemplary embodiment of beam formation for storing speaker tubes 1652 of four microphones 1656, 1658, 1660, and 1662 is shown in FIG. A first microphone 1656 is arranged closest to the tip 1654 of the speaker tube 1652, a second microphone 1658 is arranged in the speaker tube 1652 relatively farther from the tip 1654, and a third microphone 1660 is arranged in the speaker tube 1652 farther from the tip 1654. And a fourth microphone 1662 is arranged in the speaker tube 1652 farthest from the tip 1654. The first microphone 1656 and the second microphone 1658 are arranged to provide data to output a left signal 1686. The first microphone is arranged to output its signal to a gain module 1672 and a delay module 1674, which are output to a combiner 1682. The second microphone is directly connected to the combiner 1682. The combiner 1682 subtracts the two provided signals to eliminate noise, which creates a left signal 1686.
同樣地，第三麥克風1660被連接至一增益模組1676與一延遲模組1678，其會被輸出至一組合器1680。第四麥克風1662直接被連接至組合器1680。組合器1680會相減該兩個已提供的訊號，用以消除雜訊，其會創造右訊號1684。 Similarly, the third microphone 1660 is coupled to a gain module 1676 and a delay module 1678 that are output to a combiner 1680. The fourth microphone 1662 is directly connected to the combiner 1680. The combiner 1680 subtracts the two provided signals to eliminate noise, which creates a right signal 1684.
圖17所示的係接受三個訊號1710、1712、以及1714的射束形成模組1702的範例實施例的方塊圖1700。一較近麥克風訊號1710會當作主訊號1730被輸出至該ANC模組並且也會被輸入至一低通濾波器1717，以便當作主訊號1720被輸出至該VAD模組。一第一較遠麥克風訊號1712與一第二較近麥克風訊號1714會被輸入至個別的頻率響應匹配濾波器1706 與1704，它們的輸出會成為第一參考訊號1732與第二參考訊號1734被輸出至該ANC模組。該些頻率響應匹配濾波器1706與1704的輸出還會分別被輸出至低通濾波器1716與1718，它們會分別輸出第一參考訊號1722與第二參考訊號1724。 A block diagram 1700 of an exemplary embodiment of a beam forming module 1702 that accepts three signals 1710, 1712, and 1714 is shown in FIG. A closer microphone signal 1710 is output to the ANC module as the main signal 1730 and is also input to a low pass filter 1717 for output to the VAD module as the main signal 1720. A first far microphone signal 1712 and a second closer microphone signal 1714 are input to the individual frequency response matched filter 1706. And 1704, their outputs are outputted to the ANC module as the first reference signal 1732 and the second reference signal 1734. The outputs of the frequency response matched filters 1706 and 1704 are also output to low pass filters 1716 and 1718, respectively, which respectively output a first reference signal 1722 and a second reference signal 1724.
圖18所示的係接受三個訊號1820、1822、以及1824的所希望的聲頻活動偵測(VAD)模組1802的範例實施例的方塊圖1800。該VAD模組1802分別在短時功率模組1804、1805、以及1806處接收一主訊號1820、一第一參考訊號1822、以及一第二參考訊號1824。該些短時功率模組1804、1805、以及1806雷同於配合圖13所述的短時功率模組。該些短時功率模組1804、1805、以及1806輸出至個別的放大器1808、1809、以及1810，它們會各自為一對數轉換器。放大器1808與1809輸出至一組合器模組1811，其會相減該兩個訊號並且輸出差值給單通道VAD模組1814。放大器1810與1808輸出至一組合器模組1812，其會相減該兩個訊號並且輸出差值給單通道VAD模組1816。單通道VAD模組1814與1816會輸出至一邏輯或閘1818，該邏輯或閘會輸出一DVAD訊號1840。 18 is a block diagram 1800 of an exemplary embodiment of a desired audio activity detection (VAD) module 1802 that accepts three signals 1820, 1822, and 1824. The VAD module 1802 receives a main signal 1820, a first reference signal 1822, and a second reference signal 1824 at the short-term power modules 1804, 1805, and 1806, respectively. The short-term power modules 1804, 1805, and 1806 are identical to the short-time power modules described in FIG. The short time power modules 1804, 1805, and 1806 are output to individual amplifiers 1808, 1809, and 1810, which are each a pair of digital converters. Amplifiers 1808 and 1809 output to a combiner module 1811 that subtracts the two signals and outputs the difference to a single channel VAD module 1814. Amplifiers 1810 and 1808 output to a combiner module 1812 that subtracts the two signals and outputs the difference to a single channel VAD module 1816. The single channel VAD modules 1814 and 1816 are output to a logic or gate 1818 which outputs a DVAD signal 1840.
本發明的一種範例方法包含在具有麥克風陣列的眼鏡裝置中減降雜訊，其包含：耦合一麥克風陣列至一眼鏡鏡框，該麥克風陣列包含至少一第一麥克風與第二麥克風；排列該第一麥克風，用以在太陽穴區域附近耦合至該眼鏡鏡框，該太陽穴區域的位置近似介於一鏡片開口的頂端角邊與一支撐臂之間；排列該第二麥克風，用以在該鏡片開口的內側下方角邊附近耦合至該眼鏡鏡框；以及分別從該些第一麥克風與第二麥克風處提供第一音頻通道輸出與第二音頻通道輸出。該第二麥克風能夠被定位 在鏡片的內鏡框中的任何位置處，舉例來說，下方角邊、上方角邊、或是內鏡框邊緣。進一步言之，該第二麥克風能夠沿著鏡片的內緣位在鼻橋的左邊或右邊。 An exemplary method of the present invention includes reducing noise in a glasses device having a microphone array, comprising: coupling a microphone array to a spectacle frame, the microphone array including at least a first microphone and a second microphone; arranging the first a microphone for coupling to the spectacle frame near the temple region, the temple region being located approximately between a top corner edge of a lens opening and a support arm; the second microphone being arranged for inner side of the lens opening a lower corner edge coupled to the eyeglass frame; and a first audio channel output and a second audio channel output are provided from the first and second microphones, respectively. The second microphone can be positioned At any position in the inner frame of the lens, for example, the lower corner, the upper corner, or the inner frame edge. Further, the second microphone can be positioned to the left or right of the nose bridge along the inner edge of the lens.
本發明的範例方法進一步包含：在一射束形成器處形成射束，該射束形成器接收至少該些第一音頻通道與第二音頻通道並且輸出一主通道以及一或更多參考通道；在一聲頻活動偵測器處偵測聲頻活動，該聲頻活動偵測器接收該主通道與該些參考通道並且輸出一所希望的聲頻活動通道；在一適應性雜訊消除器處適應性消除雜訊，該適應性雜訊消除器接收該主通道、該些參考通道、以及該所希望的聲頻活動通道並且輸出一適應性雜訊消除通道；以及在一雜訊減降器處減降雜訊，該雜訊減降器接收該所希望的聲頻活動通道與適應性雜訊消除通道並且輸出一所希望的語音通道。於範例方法中，該些第一音頻通道與第二音頻通道會被數位式產生，並且該些射束會被數位式形成。 An exemplary method of the present invention further includes: forming a beam at a beamformer, the beamformer receiving at least the first audio channel and the second audio channel and outputting a main channel and one or more reference channels; Detecting audio activity at an audio activity detector, the audio activity detector receiving the main channel and the reference channels and outputting a desired audio activity channel; adaptively eliminating at an adaptive noise canceller Noise, the adaptive noise canceller receiving the main channel, the reference channels, and the desired audio activity channel and outputting an adaptive noise cancellation channel; and reducing noise at a noise reducer The noise reducer receives the desired audio activity channel and the adaptive noise cancellation channel and outputs a desired voice channel. In the example method, the first audio channel and the second audio channel are digitally generated, and the beams are digitally formed.
減降雜訊的範例方法會進一步包含：排列一第三麥克風，用以在該第一麥克風下方的該鏡片開口的外側下方角邊附近耦合至該眼鏡鏡框；排列一第四麥克風，用以在該第二麥克風上方的一橋接支撐區域附近耦合至該眼鏡鏡框；以及分別從該些第三麥克風與第四麥克風處提供第三音頻通道輸出與第四音頻通道輸出。 An exemplary method of reducing noise may further include: arranging a third microphone for coupling to the spectacle frame near an outer lower corner of the lens opening below the first microphone; arranging a fourth microphone for A bridge support area above the second microphone is coupled to the eyeglass frame; and a third audio channel output and a fourth audio channel output are provided from the third and fourth microphones, respectively.
根據減降雜訊的範例方法，一全向式麥克風陣列會被耦合至該眼鏡鏡框。該被耦合的全向式麥克風陣列為下面的任何組合：駐極式電容麥克風、類比式微機電系統(MEMS)麥克風、或是數位式MEMS麥克風。 According to an exemplary method of reducing noise, an omnidirectional microphone array is coupled to the spectacle frame. The coupled omnidirectional microphone array is any combination of the following: an electret condenser microphone, an analog microelectromechanical system (MEMS) microphone, or a digital MEMS microphone.
減降雜訊的範例方法會包含利用至少一撓性印刷電路板帶 將該麥克風陣列耦合至眼鏡鏡框。耦合該麥克風陣列至眼鏡鏡框會包含：利用一上撓性PCB帶，其包含第一麥克風與第四麥克風；以及利用一下撓性PCB帶，其包含第二麥克風與第三麥克風。 An example method of reducing noise would include utilizing at least one flexible printed circuit board strip The microphone array is coupled to the spectacle frame. Coupling the array of microphones to the spectacle frame can include: utilizing an upper flexible PCB strip that includes a first microphone and a fourth microphone; and utilizing a lower flexible PCB strip that includes a second microphone and a third microphone.
於本發明的範例方法中，耦合該麥克風陣列至眼鏡鏡框會進一步包含：耦合該麥克風陣列中的每一個麥克風至一出口陣列中的一對應出口，該麥克風陣列係底部埠微機電系統(MEMS)麥克風或是頂端埠微機電系統(MEMS)麥克風並且該些出口位在該眼鏡鏡框中，其中，第一MEMS麥克風與第四MEMS麥克風被耦合至上撓性PCB帶並且第二MEMS麥克風與第三MEMS麥克風被耦合至下撓性PCB帶；以及排列該MEMS麥克風陣列，使得該些埠經由該些對應出口接收聲音訊號。耦合會包含耦合一薄膜於該眼鏡鏡框與該些麥克風之間。本發明的範例方法會包含利用該薄膜為該麥克風陣列擋風與防水，該薄膜係由擋風與防水材料製成。 In an exemplary method of the present invention, coupling the microphone array to the spectacle frame further comprises: coupling each of the microphone arrays to a corresponding one of the outlet arrays, the microphone array being a bottom germanium microelectromechanical system (MEMS) a microphone or a top 埠 microelectromechanical system (MEMS) microphone and the exits are in the spectacle frame, wherein the first MEMS microphone and the fourth MEMS microphone are coupled to the upper flexible PCB strip and the second MEMS microphone and the third MEMS A microphone is coupled to the lower flexible PCB strip; and the MEMS microphone array is arranged such that the chirps receive audio signals via the corresponding outlets. The coupling may include coupling a film between the spectacle frame and the microphones. An exemplary method of the present invention would include using the film to wind and waterproof the microphone array, the film being made of a windshield and a waterproof material.
於本發明的又進一步範例實施例中，一眼鏡裝置會包含：用於耦合一麥克風陣列至一眼鏡鏡框的構件，該麥克風陣列包含至少一第一麥克風與第二麥克風；用於排列該第一麥克風的構件，使其在太陽穴區域附近耦合至該眼鏡鏡框，該太陽穴區域的位置近似介於一鏡片開口的頂端角邊與一支撐臂之間；用於排列該第二麥克風的構件，使其在該鏡片開口的內側下方角邊附近耦合至該眼鏡鏡框；以及用以分別從該些第一麥克風與第二麥克風處提供第一音頻通道輸出與第二音頻通道輸出的構件。該第二麥克風能夠被定位在鏡片的內鏡框中的任何位置處，舉例來說，下方角邊、上方角邊、或是內鏡框邊緣。進一步言之，該第二麥克風能夠沿著鏡片的內緣位在鼻橋的左邊或右邊。 In still another exemplary embodiment of the present invention, a glasses apparatus includes: means for coupling a microphone array to a spectacle frame, the microphone array including at least a first microphone and a second microphone; a member of the microphone coupled to the spectacle frame adjacent the temple region, the position of the temple region being approximately between a top corner edge of a lens opening and a support arm; a member for arranging the second microphone to Coupling to the spectacle frame near an inner lower corner of the lens opening; and means for providing a first audio channel output and a second audio channel output from the first and second microphones, respectively. The second microphone can be positioned anywhere in the inner frame of the lens, for example, a lower corner, an upper corner, or an inner frame edge. Further, the second microphone can be positioned to the left or right of the nose bridge along the inner edge of the lens.
於本發明的一又進一步範例中，用於減降雜訊的方法能夠在一非暫時性電腦程式產品上被執行，該非暫時性電腦程式產品包含一電腦可讀取的媒體，其上儲存著電腦可讀取的指令。該些電腦可讀取的指令會在被一處理器載入且執行時讓該處理器：以至少兩個音頻通道輸入為基礎形成射束並且產生一主通道及一或更多參考通道；以該主通道及該些參考通道為基礎偵測聲頻活動並且輸出一所希望的聲頻活動通道；以該主通道、該些參考通道、以及該所希望的聲頻活動通道為基礎來適應性消除雜訊並且產生一適應性雜訊消除通道；以及以該所希望的聲頻活動通道與適應性雜訊消除通道為基礎來減降雜訊並且輸出一所希望的語音通道。 In still another example of the present invention, a method for reducing noise can be performed on a non-transitory computer program product, the non-transitory computer program product comprising a computer readable medium on which is stored Computer readable instructions. The computer readable instructions, when loaded and executed by a processor, cause the processor to: form a beam based on at least two audio channel inputs and generate a primary channel and one or more reference channels; The main channel and the reference channels are based on detecting audio activity and outputting a desired audio activity channel; adaptively eliminating noise based on the main channel, the reference channels, and the desired audio activity channel And generating an adaptive noise cancellation channel; and reducing the noise based on the desired audio activity channel and the adaptive noise cancellation channel and outputting a desired voice channel.
本發明的進一步範例實施例可以利用一電腦程式產品來配置；舉例來說，可以軟體來程式化控制，用以施行本發明的範例實施例。本發明的進一步範例實施例可以包含一非暫時性電腦可讀取的媒體，其含有可被一處理器執行的指令，並且當指令被執行時可讓該處理器完成本文中所述的方法。應該瞭解的係，本文中所述之方塊圖與流程圖的元件可以軟體、硬體、韌體、或是未來決定的其它雷同施行方式來施行。此外，本文中所述之方塊圖與流程圖的元件亦可以任何方式以軟體、硬體、或韌體來組合或分割。倘若以軟體來施行，該軟體可以支援本文中揭示的範例實施例的任何語言來撰寫。該軟體可以被儲存在任何形式的電腦可讀取媒體中，例如，隨機存取記憶體(RAM)、唯讀記憶體(ROM)、唯讀光碟(CD-ROM)、「快閃(Flash)」記憶體、諸如此類。在操作中，一般用途處理器或特定應用處理器會以本技術中非常瞭解的方式載入與執行軟體。應該進一步瞭解的係，該些方塊圖與流程圖可以包含更多或較少的元件，可以不同的方式被 排列或定向，或者可以不同的方式來表現。應該瞭解的係，施行方式可以規定用以圖解本發明之實施例的執行的方塊圖、流程圖、及/或網路圖以及方塊圖與流程圖的數量。 Further exemplary embodiments of the present invention may be configured using a computer program product; for example, software may be used to programmatically control an exemplary embodiment of the present invention. Further exemplary embodiments of the present invention can include a non-transitory computer readable medium containing instructions executable by a processor and which, when executed, can cause the processor to perform the methods described herein. It should be understood that the block diagrams and flow diagram elements described herein may be implemented in software, hardware, firmware, or other similar implementations that are determined in the future. In addition, the block diagrams and elements of the flowcharts described herein may also be combined or divided in any manner by software, hardware, or firmware. If implemented in software, the software can be written in any language of the example embodiments disclosed herein. The software can be stored in any form of computer readable media, such as random access memory (RAM), read only memory (ROM), CD-ROM, and flash. "Memory, and so on. In operation, a general purpose processor or a particular application processor will load and execute software in a manner well known in the art. It should be further understood that the block diagrams and flowcharts may contain more or fewer components and may be Arrange or orient, or behave differently. It should be understood that the manner in which the embodiments may be used to illustrate the implementation of the embodiments of the present invention, FIGs, FIG.
本文以引用的方式將本文中引述的所有專利案、公開申請案、以及引證案的教示內容完整併入。 The teachings of all patents, published applications, and citations cited herein are hereby incorporated by reference in their entirety.
本文雖然已經參考本發明的範例實施例特別顯示與說明過本發明；不過，熟習本技術的人士便會瞭解，可於其中進行形式與細節的各種改變，其並沒有脫離隨附申請專利範圍涵蓋的本發明的範疇。 The present invention has been particularly shown and described with reference to the exemplary embodiments of the present invention, and it will be understood by those skilled in the art The scope of the invention.
Claims (21)
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201361839211P | 2013-06-25 | 2013-06-25 | |
| US61/839,211 | 2013-06-25 | ||
| US14/180,994 | 2014-02-14 | ||
| US14/180,994 US9753311B2 (en) | 2013-03-13 | 2014-02-14 | Eye glasses with microphone array |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201508375A TW201508375A (en) | 2015-03-01 |
| TWI624709B true TWI624709B (en) | 2018-05-21 |
Family
ID=53186189
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW103108574A TWI624709B (en) | 2013-06-25 | 2014-03-12 | Eye glasses with microphone array and method of reducing noise thereof |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI624709B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10379386B2 (en) | 2013-03-13 | 2019-08-13 | Kopin Corporation | Noise cancelling microphone apparatus |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102629385B1 (en) * | 2018-01-25 | 2024-01-25 | 삼성전자주식회사 | Application processor including low power voice trigger system with direct path for barge-in, electronic device including the same and method of operating the same |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7359504B1 (en) * | 2002-12-03 | 2008-04-15 | Plantronics, Inc. | Method and apparatus for reducing echo and noise |
| TW201006265A (en) * | 2008-07-29 | 2010-02-01 | Neovictory Technology Co Ltd | Low-background-sound bone-skin vibration microphone and glasses containing the same |
| US20110091057A1 (en) * | 2009-10-16 | 2011-04-21 | Nxp B.V. | Eyeglasses with a planar array of microphones for assisting hearing |
-
2014
- 2014-03-12 TW TW103108574A patent/TWI624709B/en active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7359504B1 (en) * | 2002-12-03 | 2008-04-15 | Plantronics, Inc. | Method and apparatus for reducing echo and noise |
| TW201006265A (en) * | 2008-07-29 | 2010-02-01 | Neovictory Technology Co Ltd | Low-background-sound bone-skin vibration microphone and glasses containing the same |
| US20110091057A1 (en) * | 2009-10-16 | 2011-04-21 | Nxp B.V. | Eyeglasses with a planar array of microphones for assisting hearing |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10379386B2 (en) | 2013-03-13 | 2019-08-13 | Kopin Corporation | Noise cancelling microphone apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201508375A (en) | 2015-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10379386B2 (en) | Noise cancelling microphone apparatus | |
| CN105765486B (en) | Wearable communication enhancement device | |
| CN103339961B (en) | For carrying out the device and method that space Sexual behavior mode sound is obtained by sound wave triangulation | |
| CN103026733B (en) | For the system of multi-microphone regioselectivity process, method, equipment and computer-readable media | |
| EP2848007B1 (en) | Noise-reducing directional microphone array | |
| AU2019203605A1 (en) | Methods circuits devices systems and associated computer executable code for acquiring acoustics signals | |
| US20080175408A1 (en) | Proximity filter | |
| US10904669B1 (en) | System for presentation of audio using wearable device | |
| JP2016516343A5 (en) | ||
| WO2020068294A1 (en) | Beamforming using an in-ear audio device | |
| CN109348322A (en) | A kind of radix saposhnikoviae method for de-noising, feed-forward noise reduction system, earphone and storage medium | |
| CN103392349A (en) | Systems, methods, apparatus, and computer-readable media for spatially selective audio augmentation | |
| US20220232342A1 (en) | Audio system for artificial reality applications | |
| TWI624709B (en) | Eye glasses with microphone array and method of reducing noise thereof | |
| WO2006132249A1 (en) | Signal separating apparatus | |
| Rahman et al. | A study on amplitude variation of bone conducted speech compared to air conducted speech | |
| US20160227320A1 (en) | Multi-channel microphone mapping | |
| TW201508376A (en) | Sound induction ear speaker for eye glasses | |
| TW201523064A (en) | Eyewear spectacle with audio speaker in the temple | |
| De Sena et al. | A generalized design method for directivity patterns of spherical microphone arrays | |
| Ogawa et al. | Speech enhancement using a square microphone array in the presence of directional and diffuse noise | |
| US11402641B1 (en) | Acoustic unit for wearable device | |
| CN116594197A (en) | Head wearing device based on three-microphone directional sound recording | |
| Vrabec et al. | MABox-Multimodal Microphone Array Algorithm Development System | |
| Miyabe et al. | Barge-in Free Spoken Dialogue Interface Based on Response Sound Cancellation Using Sound field Control and Microphone Array |