WO2013020380A1 - 一种通信耳机的语音增强方法、装置及降噪通信耳机 - Google Patents

一种通信耳机的语音增强方法、装置及降噪通信耳机 Download PDF

Info

Publication number
WO2013020380A1
WO2013020380A1 PCT/CN2012/072483 CN2012072483W WO2013020380A1 WO 2013020380 A1 WO2013020380 A1 WO 2013020380A1 CN 2012072483 W CN2012072483 W CN 2012072483W WO 2013020380 A1 WO2013020380 A1 WO 2013020380A1
Authority
WO
WIPO (PCT)
Prior art keywords
noise
signal
noise reduction
microphone
communication
Prior art date
Application number
PCT/CN2012/072483
Other languages
English (en)
French (fr)
Inventor
刘崧
李波
赵剑
Original Assignee
歌尔声学股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 歌尔声学股份有限公司 filed Critical 歌尔声学股份有限公司
Priority to JP2013552095A priority Critical patent/JP5513690B2/ja
Priority to US14/110,879 priority patent/US9484042B2/en
Priority to EP12822487.0A priority patent/EP2680608B1/en
Priority to KR1020137021521A priority patent/KR101353686B1/ko
Priority to DK12822487.0T priority patent/DK2680608T3/en
Publication of WO2013020380A1 publication Critical patent/WO2013020380A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1083—Reduction of ambient noise
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R3/00—Circuits for transducers, loudspeakers or microphones
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R1/00—Details of transducers, loudspeakers or microphones
    • H04R1/10—Earpieces; Attachments therefor ; Earphones; Monophonic headphones
    • H04R1/1041—Mechanical or electronic switches, or control elements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R2410/00—Microphones
    • H04R2410/01—Noise reduction using microphones having different directional characteristics
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R2410/00—Microphones
    • H04R2410/05—Noise reduction with a separate noise microphone
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R3/00—Circuits for transducers, loudspeakers or microphones
    • H04R3/005—Circuits for transducers, loudspeakers or microphones for combining the signals of two or more microphones

Definitions

  • the present invention relates to the field of voice enhanced noise reduction technology, and more particularly to a voice enhancement method and apparatus for providing noise reduction on a transmitting and receiving end of a communication earphone by multiplexing sound signals picked up by a plurality of microphones, and a Noise reduction communication headset.
  • the existing technical solutions perform noise reduction processing from the following two aspects: one is to use the acoustic signal processing technology to improve the signal noise of the voice signal picked up by the microphone at the communication headset. This allows the remote user to hear the speech of the communication headset user. On the other hand, it is necessary to improve the signal-to-noise ratio of the voice of the receiving end on the receiving end of the communication earphone, so that the near-end earphone wearer can hear the voice signal sent by the remote user.
  • the voice enhancement method of the communication headset of the commonly used communication headset mainly adopts single or multiple ordinary microphones to pick up signals, and then achieves the purpose of voice enhancement through the acoustic signal processing method.
  • Single microphone speech enhancement is generally referred to as single-channel spectral subtraction speech enhancement technology (see Chinese Patent Application Publication No. CN1684143A, CN101477800A).
  • This technique generally estimates the energy of steady-state noise in the current speech by analyzing historical data, and then The method of spectral subtraction eliminates the noise in the speech to achieve the purpose of speech enhancement, but this method can only suppress the steady-state noise (such as white noise), and the amount of noise reduction is limited. If the amount of noise reduction is too large, the speech will be damaged. For non-stationary noise (such as surrounding speech noise, knocking noise), its energy cannot be accurately estimated and therefore cannot be effectively eliminated.
  • Another method for effectively suppressing unsteady noise is to use a microphone array speech enhancement technique composed of two or more microphones (see Chinese Patent Application Publication No. CN101466055A, CN1967158A), which is usually received by a microphone.
  • the obtained signal is used as a reference signal, and the adaptive filtering method is used to estimate and cancel the noise component in the other microphone pickup signal in real time, and the speech component is preserved, thereby achieving the purpose of speech enhancement.
  • the multi-microphone technology can suppress the unsteady noise and reduce the noise. The amount is greater than the single microphone technology.
  • this method requires accurate detection of the speech state, otherwise the speech may be eliminated as noise.
  • Some of the existing multi-microphone technologies use a directional microphone (see Chinese Patent Application Publication No. CN101466055A) or a plurality of microphones to form directivity (see Chinese Patent Application Publication No. CN101466056A) for detecting speech from a specific direction.
  • the method is only applicable when the shape of the microphone array is fixed and fixed relative to the user. When the user deviates from the pointing range of the microphone array, or the position of the microphone array changes, causing the orientation of the microphone array to deviate from the user, the speech is treated as noise suppression.
  • the case where the microphone shown in Fig. 1 is mounted on the earphone cord.
  • the microphone 112 is disposed on the earphone cord, and in a specific application process, the earphone microphone is not fixed with respect to the user's mouth, and the microphone is disposed on other parts of the earphone.
  • the shape of the microphone array is not fixed.
  • the microphone will take the microphone on the cord to any position on the mouth.
  • the voice will be regarded as noise. To handle, at this time, the directivity of the microphone array cannot accurately detect the voice.
  • the voice enhancement method of the communication terminal of the commonly used communication earphone mainly adopts two technologies.
  • One is to adopt an automatic volume control technology (see Chinese Invention Patent Application Publication No. CN1507293A), which automatically increases the power output to the speaker unit when the external noise is high, which is a passive method limited by the power of the speaker unit itself.
  • In-ear earphones are fed into the industry standard for human ear sound pressure.
  • the volume of the speaker unit cannot be increased without limit, and the high-intensity voice emitted by the speaker can also cause damage to the user's own hearing and physical and mental health.
  • Another method is to apply a traditional active/passive noise control technology (see Chinese invention patent application specification CN101432798A) to a communication earphone.
  • the earphone is divided into a headphone type and an earphone type, and the earphone type earphone is generally used.
  • the holster and the human ear are sealed in a sealed form.
  • the medium and high frequency noise are reduced by the sound absorption and sound insulation of the material, and on the other hand, the active noise control technology is effective to reduce the low frequency (mainly below 300 Hz) noise. Therefore, the control effect of the external frequency noise on the whole frequency band is better, and the signal-to-noise ratio of the voice of the communication terminal of the communication earphone can be effectively improved.
  • the speech enhancement technology of the sending end recognizes the wearing state of the earphone according to the difference of the energy of the voice signals picked up by the plurality of microphones, thereby selecting different noise reduction processing methods to ensure that the voice is not damaged regardless of how the earphone is worn, and When the earphone is worn normally, it can provide a good noise reduction effect.
  • the receiver uses non-closed feedforward active noise control technology to reduce the noise while ensuring the comfort of the headphones.
  • a voice enhancement method for a communication headset comprising a transmitter terminal composed of at least two microphones and a receiver terminal composed of at least one microphone and a speaker, the method
  • the noise reduction process is respectively provided on the sending end and the receiving end of the communication earphone by using a plurality of microphone signals, wherein the noise reduction processing at the sending end includes:
  • Determining a wearing state of the communication earphone by comparing an energy difference of a sound signal picked up by a microphone of the communication earphone; if the energy difference is greater than a first preset threshold, determining that the communication earphone is normally worn, first speaking the sound
  • the signal is subjected to multi-microphone noise reduction, and then the residual steady-state noise is further suppressed by single-channel noise reduction; otherwise, the communication earphone is determined to be abnormally worn, and the steady-state noise in the sound signal is directly suppressed by single-channel noise reduction.
  • the method specifically includes: distinguishing a voice signal component in the sound signal by comparing energy differences of frequency components in the sound signal. a noise signal component; attenuating the noise signal component.
  • a communication earphone comprising a transmitting end composed of at least two microphones and a receiving end composed of at least one microphone and one speaker, and a noise reduction at the transmitting end
  • a wearing state determining module configured to determine a wearing state of the communication earphone by comparing an energy difference of a sound signal picked up by a microphone constituting the sending end, and if the energy difference is greater than a first preset threshold, determining that the communication earphone is normally worn, Otherwise, determining that the communication headset is not properly worn;
  • a multi-microphone noise reduction module configured to perform multi-microphone noise reduction processing on the sound signal when the communication earphone is normally worn
  • a single-channel noise reduction module configured to further suppress residual steady-state noise after the noise reduction processing of the sound signal by the multi-microphone noise reduction module, and when the communication earphone is abnormally worn, directly The steady-state noise in the sound signal is suppressed.
  • a voice enhancement apparatus including a voiceband noise reduction processing unit and a voice receiving noise reduction processing unit, wherein the voiceband noise reduction processing unit includes:
  • a voice-sending mode determining module configured to determine a noise reduction mode of the sending end by comparing energy differences of sound signals picked up by a microphone constituting the sending end;
  • a multi-microphone noise reduction module configured to perform multi-microphone noise reduction processing on the sound signal when the energy difference is greater than a first preset threshold
  • a single-channel noise reduction module configured to further suppress residual steady-state noise after the multi-microphone noise reduction module performs noise reduction processing on the sound signal, and wherein the energy difference is less than or equal to the first preset
  • the threshold directly suppresses the steady-state noise in the sound signal.
  • the earplug of the present invention is designed as a non-closed in-ear structure to ensure long-term wearing comfort, and a feedforward active noise control technique is implemented on the non-closed earphone to achieve noise reduction in the voice band. To ensure a high signal to noise ratio of the voice at the receiving end.
  • the howling detection unit is further added to adjust the noise reduction processing mode of the receiving end in time by changing the sound signal picked up by the transmitting end, thereby increasing the robustness of the system.
  • the communication earphone and the voice enhancement device of the communication earphone By using the voice enhancement method, the communication earphone and the voice enhancement device of the communication earphone according to the present invention, the signals picked up by the plurality of microphones can be effectively multiplexed, and the voice signal enhancement method is adopted in the communication earphone sending and receiving ends to implement voice enhancement. To ensure high signal-to-noise ratio of near-end and far-end speech in noisy environments, high-definition and intelligible speech signals are provided for both parties.
  • FIG. 1 is a schematic view showing the assembly structure of a microphone on a communication earphone in the prior art
  • FIG. 2 is a schematic structural diagram of a communication earphone according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of a system of a communication earphone according to an embodiment of the present invention.
  • FIG. 4 is a flow chart showing a portion of a voice canceling processing portion of a voice transmitting method of a communication earphone according to the present invention
  • FIG. 5 is a schematic diagram showing the logical structure of a noise reduction processing unit of a transmitting end according to an embodiment of the present invention
  • FIG. 6 is a flow chart showing a portion of a speech side noise reduction processing in a speech enhancement method of a communication earphone according to the present invention
  • FIG. 7 is a schematic diagram showing the logical structure of a receiver-side noise reduction processing unit according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of a normal wearing state of an earphone according to an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of an abnormal wearing state of an earphone according to an embodiment of the invention.
  • the noise is effectively attenuated and suppressed on the basis of not damaging the voice signal, and the present invention simultaneously performs noise reduction from the transmitting end and the receiving end, and according to multiple microphones.
  • the specific characteristics of the received sound signal are mainly the difference between the component energy of the speech signal and the noise signal contained therein to identify the wearing state of the earphone, and the corresponding speech enhancement and noise reduction methods are adopted, so that the noise reduction processing is more targeted. To ensure voice quality and achieve better noise reduction.
  • the key of the voice enhancement method of the communication earphone provided by the invention is to effectively multiplex the sound signal collected by the microphone array, and adopt multi-microphone voice enhancement technology and non-closed feedforward active noise control respectively at the transmitting and receiving ends of the communication earphone.
  • the technology is used to improve the signal-to-noise ratio of the voice and the voice of the communication headset in a noisy environment, and to ensure the clarity and intelligibility of the communication voice.
  • the present invention proposes a multi-mike noise reduction technology combined with user wearing state recognition at the transmitting end, which does not need to use microphone directivity to detect speech, but detects the main signal and the reference signal in the sound signal picked up by the microphone.
  • the energy difference to identify different user wearing states corresponds to different noise reduction methods, so as to ensure that the noise reduction process does not damage the voice when the microphone position or shape is not fixed.
  • the present invention adopts a non-closed feedforward active noise control technology to ensure wear comfort while effectively reducing noise signals in the voice band.
  • the voice enhancement method of the communication earphone provided by the invention performs noise reduction processing from both the sending end and the receiving end. Since the present invention performs noise reduction processing on the basis of the sound signal collected by the multiplexed microphone, the present invention
  • the communication earphone to which the invention is applied includes a transmitting end composed of at least two microphones, a receiving end composed of at least one microphone and one speaker, and a host for performing noise reduction processing on the sound signal.
  • 2 is a block diagram showing an embodiment of a communication earphone to which the present invention is applied.
  • the ear portion of the communication earphone used in this embodiment adopts a conventional non-closed in-ear earphone structure, which can be well coupled with the ear to provide a firmness of wearing without completely sealing the ear canal. Long-wearing comfort.
  • the communication earphone comprises a sending end, a receiving end, a headphone line and a host 230. wherein the sending end uses the signals collected by the three microphones, the microphone 212 is disposed on the headphone line, and the microphones 214 and 216 are respectively disposed on the headphone strut On the back, the opening is facing outwards.
  • the receiver end includes two microphones 214 and 216 and two speakers 224 and 226.
  • the user can take the microphone 212 mounted on the earphone line to the mouth when the earphone is normally worn (as shown in FIG. 8), since the microphone 212 is close to the mouth, it can pick up
  • the signal to noise ratio is relatively high, so the microphone 212 is considered to be the main microphone.
  • the microphones 214 and 216 are disposed on the back of the earphone pole, and the opening is outward. During the normal use of the communication earphone, the distance from the user's mouth is far away, so that it is easy to pick up a good noise reference signal.
  • a microphone is considered a reference microphone.
  • FIG. 3 a system block diagram of a three-microphone communication headset 300 is shown in FIG. 3, wherein the host side includes a DSP unit 200 and a receiver terminal voice noise reduction processing unit 700, and a DSP portion is sent.
  • the voice reduction processing unit 400 completes the enhancement of the voice of the voice transmission terminal, and the howling detection unit 500 provides the howling detection control signal for the voice enhancement module of the receiving end; the voice reduction processing unit 700 of the receiving end completes the voice drop of the receiving end. noise.
  • the host side can be implemented by using DSP plus some analog circuits, or as part of some audio equipment or mobile phones.
  • the number of microphones used in the embodiment shown in FIG. 3 is three, but other numbers of microphones may be used in the specific application of the present invention, such as only being disposed on the back of the earphone support.
  • Two microphones (such as 214 and 216), there is no difference between the main microphone and the reference microphone at the sending end, only the single-channel noise reduction mode can be used; if two microphones are arranged on the back of the headphone cable and the headphone rod (such as 212 and 214), you can choose multi-microphone noise reduction mode and / or single-channel noise reduction mode according to the user's wearing situation; or use more microphones according to the needs of specific communication products to achieve better pickup of useful speech.
  • the signal and the noise signal it is possible to determine whether there is a distinction between the master and the slave microphone according to the sound signal picked up by the microphone, thereby adopting the corresponding noise reduction mode.
  • Fig. 4 is a flow chart showing a portion of the noise reduction processing of the transmitting terminal in the voice enhancement method of the communication earphone according to the present invention.
  • the noise reduction processing procedure of the noise reduction processing part of the sending end includes:
  • S410 determining a difference in energy of the microphone pickup signal by comparing energy of a sound signal picked up by a microphone of the receiving end of the communication earphone, wherein the sound signal comprises a voice signal and a noise signal;
  • S420 Determine whether the determined energy difference is greater than the first preset threshold to identify the headset wearing state. If the energy difference is greater than the first preset threshold, the headset is normally worn (as shown in FIG. 8), then proceed to step S430, otherwise, the headset Abnormal wear (as shown in Figure 9), proceeds to step S440;
  • FIG. 5 is a schematic diagram showing the logical structure of a speech-side noise reduction processing unit for performing voice enhancement by using an acoustic signal processing method at a communication headset transmitting end according to an embodiment of the invention.
  • the transmitting end noise reduction processing unit 400 includes a wearing state determining module 420, a multi-microphone noise reduction module 440, and a single channel noise reduction module 460.
  • the wearing state determining unit 420 is configured to determine a wearing state of the communication earphone by comparing the energy difference of the sound signal picked up by the microphone constituting the sending end, and if the energy difference is greater than the first preset threshold, determining that the communication earphone is normally worn, otherwise, It is determined that the communication headset is not properly worn.
  • the picked up sound signal comprises a voice signal and a noise signal;
  • the multi-microphone noise reduction module 440 is configured to perform multi-microphone noise reduction processing on the sound signal picked up by the microphone when the energy difference is greater than the first preset threshold and the communication headset is normally worn;
  • the single-channel noise reduction module 460 is configured to further suppress residual steady-state noise after the multi-microphone noise reduction module 440 performs noise reduction processing on the sound signal, and, when the energy difference is less than or equal to the first preset threshold, and the communication earphone is in a non- In the normal wearing state, the steady-state noise in the sound signal is directly suppressed.
  • noise reduction processing method and the noise reduction processing module of the present invention at the transmitting end will be described in more detail below with reference to FIG. 3, FIG. 4 and FIG.
  • the distance and position of the microphones 214, 216 with respect to the human mouth are substantially determined, and are regarded as reference microphones in the present invention, and the sound signals picked up by them are regarded as reference signals.
  • the microphone 212 is brought to a position close to the human mouth, which is regarded as the main microphone in the present invention, and the picked-up sound signal is regarded as the main signal.
  • the position of the microphone 212 has a large uncertainty in actual use. It may be closer to the mouth of the person or to the distance of the microphones 214, 216 to the mouth of the person.
  • the microphone 212 is defined as a normal wearing mode when the distance between the microphone 212 and the human mouth is relatively close.
  • the main signal picked up by the microphone 212 is higher than the reference signal picked up by the microphones 214 and 216, and is generally in the normal communication environment.
  • the signal will be more than 6 dB higher than the reference signal; and when the microphone 212 is too far from the human mouth, it is in an abnormal wearing mode, at which time the microphone 212
  • the picked up main signal energy is close to the reference signal energy picked up by the microphones 214, 216.
  • the energy and P_112, P_114; then the ratio of energy sum Rp P_112 / P_114.
  • Rp is greater than the threshold Rth (for example, Rth>6dB)
  • the multi-microphone noise reduction unit 460 performs multi-macker noise reduction processing on the sound signal, and then performs single-channel noise reduction.
  • Rp is smaller than the threshold Rth, it is an abnormal wearing mode.
  • the speech and noise cannot be distinguished very well. If the multi-mike noise reduction is also used, the speech may be suppressed as noise, so only the single-channel noise reduction unit 480 is used. Noise processing to avoid speech damage.
  • the multi-microphone noise reduction module 440 includes a sound signal component discrimination module 442 and a noise signal attenuation module 444.
  • the sound signal component distinguishing module 442 is configured to distinguish the voice signal component and the noise signal component in the sound signal by comparing the energy difference of each frequency component in the sound signal; the noise signal attenuation module 444 is configured to distinguish the sound signal component distinguishing module 442 The noise signal component is attenuated.
  • the multi-microphone noise reduction unit 460 distinguishes the speech component and the noise component by the energy difference of each frequency component in the signal picked up by the microphone 212 and the microphone 214 (that is, the main microphone and the reference microphone), and performs noise reduction processing on the noise component.
  • the voice signal component discrimination module 442 is used to distinguish the voice signal from the noise signal.
  • the specific processing includes:
  • the speech component is then maintained and the noise component is attenuated by the noise signal attenuation module 444. That is, when Ri is greater than the threshold Rthi (Rthi>6dB), no processing is performed for Fi_112; when Ri is less than the threshold Rthi (Rthi>6dB), Fi_112 is multiplied by a gain Gi (0 ⁇ Gi ⁇ 1), thereby achieving Noise reduction effect.
  • the processed Fi_112 is inverse Fourier transformed to obtain a pure speech signal after noise reduction.
  • the principle of noise reduction of the single-channel noise reduction module 460 in the present invention is to calculate the energy of the stationary noise in each frequency band of the input signal and eliminate it according to the characteristics of noise smoothness.
  • the single-channel noise reduction module 460 includes a noise energy statistics module 462 and a noise energy removal module 464, wherein the noise energy statistics module 462 is configured to calculate the frequencies of the sound signal by using a smooth average method.
  • the noise energy removing module 464 is configured to remove the noise energy calculated by the noise energy statistic module 462 in the sound signal, thereby further reducing the noise component and retaining the voice component, so as to improve the signal-to-noise ratio of the voice signal.
  • the invention adopts a feedforward active noise control method to perform noise reduction processing on the receiving end.
  • the earphone part of the communication earphone adopts a non-closed earplug structure, which is mainly to ensure that the air pressure in the ear canal is consistent after wearing the earphone, and the comfort of long-term wearing is ensured.
  • the feedforward active noise control microphone is generally placed on the outer surface of the communication earphone, and as much as possible to pick up the information of the external noise. Therefore, the structure of such feedforward active noise control applied to the communication earphone generally satisfies the system's requirement for causality.
  • the sound transmitted from the front of the microphone must first reach the microphone and then reach the human ear from other directions.
  • the noise passed is basically picked up by the microphone because it is going through the diffraction of the human head.
  • Fig. 6 is a flow chart showing the portion of the speech side noise reduction processing in the speech enhancement method of the communication earphone according to the present invention.
  • the process for reducing the noise signal in the received voice band by using the feedforward active noise control method on the receiving end includes:
  • S610 picking up a noise signal by using a microphone of the receiving end of the communication headset
  • S620 Determine an anti-noise signal according to the picked-up noise signal
  • the noise signal is first inverted by the inverter to determine the primary anti-noise signal; and then the phase compensator is used in the audio range for the primary anti-noise signal.
  • the phase is corrected and adjusted to determine an anti-noise signal that is completely inverted in phase of the noise signal, and an active filter including a dual T-type network is used to compensate for the low frequency phase of the low frequency portion due to the non-closed structure loss.
  • FIG. 7 is a schematic diagram showing the logical structure of a receiver-side noise reduction processing unit according to an embodiment of the present invention.
  • the receiver-side noise reduction processing unit 700 includes a noise signal determination module 720, an anti-noise signal determination module 740, and an output signal mixing module 760, wherein the anti-noise signal determination module 740 can include an inverter 743 and a phase. Compensator 744.
  • the noise signal picking module 720 is configured to pick up a noise signal by using a microphone of the receiving end of the communication earphone. Since the receiving end receives the far-end voice signal, the sound signal picked up by the microphone is generally regarded as a noise signal, and therefore, is disposed on the earphone support rod.
  • the microphones 214 and 216 on the back are also equivalent to the noise signal pickup module 720 herein.
  • the anti-noise signal determining module 740 is configured to determine an anti-noise signal according to the noise signal determined by the noise signal determining module 720; the output signal mixing module 760 is configured to use the anti-noise signal determined by the anti-noise signal determining module 740 with the receiving end
  • the received speech signals are mixed and fed into the human ear through the speaker 224 constituting the receiving end, and the anti-noise and the original noise of the ear (transmitted by the natural acoustic channel) cancel each other and the speech signal does not change, thereby reducing the received speech band. Noise signal.
  • the inverter 742 is configured to perform inverse processing on the noise signal to determine a primary anti-noise signal
  • the phase compensator 744 is configured to correct and adjust the phase of the primary anti-noise signal in the audio range, determine an anti-noise signal that is completely inverted from the phase of the noise signal, and employ an active filter including a dual T-type network. Used to compensate for low frequency phase losses due to non-closed structures in the low frequency portion.
  • the receiver-side noise reduction processing unit 700 may further include a first amplifier 730 for amplifying the noise signal picked up by the noise signal pickup module 720 and a second amplifier 750 for using the second amplifier 750 for A mixed signal mixed with an anti-noise signal and a voice signal is amplified.
  • the noise signal picked up by the microphone 214 is amplified by the pre-first amplifier 730 and then passed through an inverter 742 and a phase compensator 744 to generate an anti-noise signal of the same phase as the original noise amplitude.
  • the phase compensator 744 is mainly for solving the delay problem of the feedforward active noise control technology applied to the non-closed communication earphone, and the phase of the anti-noise signal is correspondingly corrected and adjusted in the audio range from the circuit, in order to achieve The anti-noise and original noise phases are just reversed for the purpose. Its general implementation is implemented using a passive or active dual T-type network.
  • the anti-noise signal and the input speech signal are mixed on the circuit via an output signal mixing module 760 formed by an adder as an input to the back end second amplifier 750, which mixes the mixed signal of the anti-noise and the speech signal. Zoom in and push the speaker 224 directly.
  • the noise signal picked up by the microphone 216 is amplified by the pre-first amplifier 730, inverted by the inverter 742, compensated by the phase compensator 744, and mixed by the adder, amplified by the second amplifier 750, and directly pushed to the speaker 226.
  • the mic preamplifier first amplifier 730, the inverter 742, the phase compensator 744, the adder, the speaker power second amplifier 750, etc. may be implemented by separate devices or the same device may implement one or several modules.
  • a mixed signal mixed with anti-noise and speech passes through the speaker, and the sound signal is fed into the human ear.
  • the anti-noise signal emitted from the speaker and the original noise signal transmitted from the acoustic channel to the human ear are in the same phase, so they will be in the opposite direction.
  • the superimposed offset occurs at the ear, so as to eliminate the original noise and anti-noise, the noise is reduced, and the speech energy is unchanged, which effectively improves the signal-to-noise ratio of the speech signal, and the transmission to the human ear will be clear. Understand the purer voice signal.
  • the external noise needs to pass through the passive sound insulation material from the reference microphone to the human ear, which will increase the delay of the acoustic channel, thus providing more processing time for the electronic channel.
  • the causality of the system In order to solve the delay problem of the feedforward active noise control technology applied to the non-enclosed structure communication earphone, it is necessary to select and design the system from two aspects. First, it is necessary to better the front and back cavity of the speaker unit. Design and process, adjust the size and opening of the front and rear cavity to improve the phase response between the speaker and the human ear in the audio range, and secondly, phase compensation the inverter on the circuit, and correct the delay from the circuit itself. And compensation, in order to have better noise reduction effect in the entire audio range.
  • the microphone and the human ear are required to have a certain distance to ensure that there is more processing time on the electronic level during the time when the noise propagates from the microphone to the human ear.
  • the microphone and the speaker also need to maintain a certain spatial distance and better acoustic isolation to prevent the signal emitted by the speaker from being picked up by the microphone, thereby avoiding the noise signal picked up by the microphone including useful voice signals and causing the entire system.
  • a feedback loop that forms a feedback howling is formed. If there is a feedback loop, if the system gain is too high, there may be a howling phenomenon.
  • non-closed feedforward active noise canceling headphones there is an inherent leak path between the speaker and the reference microphone that picks up external noise.
  • the amplitude of the acoustic transfer function from the speaker to the reference microphone is very small, so in normal use, the non-closed feedforward active noise control technology will not damage the voice signal, and the system will not be unstable. The howling phenomenon.
  • the amplitude of the acoustic transfer function between the speaker and the reference microphone will increase dramatically, especially in the high frequency portion.
  • This large-scale acoustic transfer function and the control circuit with high gain constitute a closed-loop feedback system.
  • the system When the amplitude and phase of the closed-loop feedback system meet certain conditions, the system will generate self-excited howling, which is robust. Sexual problems.
  • the DSP unit further includes a howling detection mode unit for providing a howling detection control signal for the voice receiving module of the receiving end, specifically, the microphone of the communication earphone is picked up.
  • a howling detection mode unit for providing a howling detection control signal for the voice receiving module of the receiving end, specifically, the microphone of the communication earphone is picked up.
  • control circuit can be implemented by reducing the gain of the first amplifier or directly cutting off the power of the active noise control circuit.
  • the voice signal of the sending end and the receiving end it can be connected with other devices in a wired manner, or can be connected with other devices by using a wireless connection method such as Bluetooth.

Abstract

本发明提供了一种通信耳机的语音增强方法，包括送话端降噪处理和受话端降噪处理两部分，其中，送话端降噪处理部分包括：通过比较通信耳机的麦克风拾取的声音信号的能量差异确定耳机的佩戴状态；如果耳机正常佩戴，则先对声音信号进行多麦克风降噪，然后通过单通道降噪进一步抑制残留的稳态噪声；否则，直接通过单通道降噪抑制声音信号中的稳态噪声。本发明有效复用多个麦克风拾取到的信号，同时在通信耳机送、受话端分别采用声学信号处理和主动噪声控制方法实现语音增强，保证噪声环境下，近端和远端语音的高信噪比，为通信双方提供高清晰度和可懂度的语音信号。

Description

一种通信耳机的语音增强方法、装置及降噪通信耳机
本发明涉及语音增强降噪技术领域，更为具体地，涉及一种通过复用多个麦克风所拾取的声音信号在通信耳机的送、受话端提供降噪的语音增强方法和装置，以及一种降噪通信耳机。
社会信息化程度的提高使得人们能随时随地的进行通信和交流，各种各样的通信设备和技术的广泛应用极大地方便了人们的生活和提高了工作效率。然而社会的发展带来的一个比较严重的问题是噪声问题，在噪声环境中进行通信，严重影响到通信语音的清晰度和可懂度，当噪声高到一定程度时，不但通信无法进行，而且会对人的听力和身心健康产生伤害。
针对在强噪声背景下进行通信问题，现有的技术解决方案从以下两个方面来进行降噪处理：一方面是在通信耳机送话端采用声学信号处理技术提高麦克风拾取的语音信号的信噪比，使得远端用户能够听清通信耳机使用者的讲话。另一方面需要在通信耳机受话端提高受话端语音的信噪比，使得近端耳机佩戴者能够听清远端用户送过来的语音信号。
目前常用的通信耳机送话端的语音增强方法主要是采用单个或多个普通麦克风拾取信号，然后通过声学信号处理方法来达到语音增强的目的。
单个麦克风语音增强一般称之为单通道谱减语音增强技术（参见中国发明专利申请公开说明书CN1684143A，CN101477800A），这种技术一般通过对历史数据的分析来估计当前语音中稳态噪声的能量，然后通过频谱相减的方法消除语音中的噪声来达到语音增强的目的，但是这种方法只能抑制稳态的噪声（如白噪声），而且降噪量有限，降噪量过大就会损害语音，对于非稳态噪声（如周围的语音噪声，敲击噪声）不能准确估计其能量，因此不能被有效消除。
另一种可以有效的抑制非稳态噪声的方法是采用两个或多个麦克风组成的麦克风阵列语音增强技术（参见中国发明专利申请公开说明书CN101466055A，CN1967158A），这种技术通常是用一个麦克风接收到的信号作为参考信号，通过自适应滤波的方法实时估计并抵消另外一个麦克风拾取信号中的噪声成分，保留语音成分，从而达到语音增强的目的，多麦克风技术可以抑制非稳态噪声而且降噪量大于单麦克风技术。但是这种方法需要准确检测语音状态，否则语音有可能会被当作噪声被消除。
现有的一些多麦克风技术是利用指向性麦克风（参见中国发明专利申请公开说明书CN101466055A）或多个麦克风形成指向性（参见中国发明专利申请公开说明书CN101466056A）对来自特定方向的语音进行检测，这种方法只适用于麦克风阵列形状固定、且与使用者的相对位置或方向固定的情况。当用户偏离麦克风阵列的指向范围，或麦克风阵列形状位置发生变化导致麦克风阵列的指向偏离用户时，语音就会被当作噪声抑制。如图1所示的麦克风装配在耳机软线上的情形。
在图1所示的通信耳机中，麦克风112设置在耳机软线上，在具体应用过程中，这种耳机麦克风相对于使用者的嘴是不固定的，并且与设置在耳机上其他部位的麦克风所构成的麦克风阵列的形状也不固定，通话时使用者会将软线上的麦克风拿到嘴边的任意位置，当用户把麦克风拿到麦克风阵列的指向性范围外，语音就会被当成噪声来处理，此时利用麦克风阵列的指向性就不能准确检测到语音。
目前常用的通信耳机受话端的语音增强方法主要采用两种技术。一种是采用自动音量控制技术（参见中国发明专利申请公开说明书CN1507293A），即外界噪声高的时候自动提高输出给扬声器单元的功率，这是一种被动的方法，受限于扬声器单元本身的功率及入耳式耳塞馈入人耳声压的行业标准，扬声器单元的音量不可能无限制的提高，另外扬声器发出的高强度的语音对使用者本身的听力和身心健康也会产生伤害。另外一种方法是把传统的主动/被动相结合的噪声控制技术（参见中国发明专利申请公开说明书CN101432798A）应用于通信耳机，这种耳机分头戴式和耳塞型两种，耳塞型耳机一般采用皮套和人耳进行密封性的耦合形式，一方面通过材料的吸声和隔声来降低中、高频噪声，另一方面通过主动噪声控制技术有效的降低低频（主要在300Hz以下）噪声，从而实现在全频带对外界噪声较好的控制效果，可以较有效的提升通信耳机受话端语音的信噪比。
 技术问题
但是长期佩戴密封式的耳塞型通信耳机，使用者会有耳道内外气压不均衡的感觉，因此，佩戴的舒适性是制约这种结构的主动降噪技术不能广泛应用于通信耳机的主要原因。
另外，在强噪声环境下进行通信，需要同时对送、受话端语音进行降噪增强（参见中国发明专利CN101853667A），这种分别在送话端提供自适应滤波加单通道降噪和在受话端提供封闭式反馈主动降噪来实现通信双方的语音增强技术，除了分别在送、受话端存在前述提到的应用局限性之外，由于其近端自适应滤波的噪声参考信号取自受话端的封闭式反馈主动噪声控制系统，也存在无法保证噪声相关性和因果性的问题。
 技术解决方案
鉴于上述问题，本发明的目的是提供一种复用多个麦克风所采集到的信号来进行语音增强和降噪的技术。其中送话端的语音增强技术根据多个麦克风拾取到的语音信号能量差异来识别耳机的佩戴状态，从而选择不同的降噪处理方法，以保证无论耳机如何佩戴，语音都不会受到损伤，而且在耳机正常佩戴时，可以提供很好的降噪效果。而受话端采用非封闭式前馈主动噪声控制技术，降噪的同时保证了耳机佩戴的舒适性。
根据本发明的一个方面，提供了一种通信耳机的语音增强方法，该通信耳机包括由至少两个麦克风构成的送话端和由至少一个麦克风及一个扬声器构成的受话端，所述方法复用多个麦克风信号在所述通信耳机的送话端和受话端分别提供降噪处理，其中，在所述送话端的降噪处理包括：
通过比较所述通信耳机的麦克风拾取的声音信号的能量差异确定所述通信耳机的佩戴状态；如果所述能量差异大于第一预置阈值，确定所述通信耳机正常佩戴，则先对所述声音信号进行多麦克风降噪，然后通过单通道降噪进一步抑制残留的稳态噪声；否则，确定所述通信耳机非正常佩戴，直接通过单通道降噪抑制所述声音信号中的稳态噪声。
其中，优选的方案是，在对所述声音信号进行多麦克风降噪的过程中，具体包括：通过比较所述声音信号中各频率成分的能量差异来区分所述声音信号中的语音信号成分和噪声信号成分；对所述噪声信号成分进行衰减处理。
根据本发明的另一方面，提供了一种通信耳机，所述通信耳机包括由至少两个麦克风构成的送话端和由至少一个麦克风和一个扬声器构成的受话端，以及送话端降噪处理单元和受话端降噪处理单元，其中，所述送话端降噪处理单元包括：
佩戴状态确定模块，用于通过比较构成送话端的麦克风拾取的声音信号的能量差异确定所述通信耳机的佩戴状态，如果所述能量差异大于第一预置阈值，确定所述通信耳机正常佩戴，否则，确定所述通信耳机非正常佩戴；
多麦克风降噪模块，用于在所述通信耳机正常佩戴时，对所述声音信号进行多麦克风降噪处理；
单通道降噪模块，用于在所述多麦克风降噪模块对所述声音信号进行降噪处理后，进一步抑制残留的稳态噪声，以及，在所述通信耳机非正常佩戴时，直接对所述声音信号中的稳态噪声进行抑制处理。
根据本发明的另一方面，提供了一种语音增强装置，包括送话端降噪处理单元和受话端降噪处理单元，其中，所述送话端降噪处理单元包括：
送话端降噪模式确定模块，用于通过比较构成所述送话端的麦克风拾取的声音信号的能量差异确定所述送话端的降噪模式；
多麦克风降噪模块，用于在所述能量差异大于第一预置阈值时，对所述声音信号进行多麦克风降噪处理；
单通道降噪模块，用于在所述多麦克风降噪模块对所述声音信号进行降噪处理后，进一步抑制残留的稳态噪声，以及，在所述能量差异小于等于所述第一预置阈值，直接对所述声音信号中的稳态噪声进行抑制处理。
另外，在受话端，本发明的耳塞设计为非封闭式的入耳结构，以保证长期佩戴的舒适性，同时在非封闭式耳机上实现前馈主动噪声控制技术，在语音频带上实现降噪，保证受话端语音的高信噪比。
在本发明的一个优选实施方式中，还增加了啸叫检测单元以通过对送话端所拾取的声音信号的变化及时调整受话端的降噪处理方式，从而增加系统的鲁棒性。
利用上述根据本发明的通信耳机的语音增强方法、通信耳机以及语音增强装置，可以有效复用多个麦克风拾取到的信号，同时在通信耳机送、受话端采用声学信号处理方法实现语音增强，保证噪声环境下，近端和远端语音的高信噪比，为通信双方提供高清晰度和可懂度的语音信号。
为了实现上述以及相关目的，本发明的一个或多个方面包括后面将详细说明并在权利要求中特别指出的特征。下面的说明以及附图详细说明了本发明的某些示例性方面。然而，这些方面指示的仅仅是可使用本发明的原理的各种方式中的一些方式。此外，本发明旨在包括所有这些方面以及它们的等同物。
 附图说明
通过参考以下结合附图的说明及权利要求书的内容，并且随着对本发明的更全面理解，本发明的其它目的及结果将更加明白及易于理解。在附图中：
图1为现有技术中麦克风在通信耳机上的装配结构示意图；
图2为根据本发明实施例的通信耳机的结构示意图；
图3为本发明实施例的通信耳机的系统结构示意图；
图4为根据本发明的通信耳机的语音增强方法中送话端降噪处理部分的流程图；
图5为根据本发明实施例的送话端降噪处理单元的逻辑结构示意图；
图6为根据本发明的通信耳机的语音增强方法中受话端降噪处理部分的流程图；
图7为根据本发明实施例的受话端降噪处理单元的逻辑结构示意图；
图8为根据本发明实施例的耳机正常佩戴状态示意图；
图9为根据本发明实施例的耳机非正常佩戴状态示意图。
在所有附图中相同的标号指示相似或相应的特征或功能。
 本发明的实施方式
为了克服现有技术在降噪方案上所存在的不足，在不损害语音信号的基础上有效地衰减和抑制噪声，本发明从送话端和受话端同时进行降噪，并根据多个麦克风所接收的声音信号的具体特点，主要是其中所包含的语音信号和噪声信号的成分能量差异来识别耳机的佩戴状态，并采用相应的语音增强、降噪方法，使得降噪处理更有针对性，保证语音质量，并达到更好的降噪。
下面以常用的通信耳机为例来对本发明提出的语音增强方法流程和装置的结构进行详细说明。
本发明所提供的通信耳机的语音增强方法的关键在于有效复用麦克风阵列所采集到的声音信号，在通信耳机送、受话端分别采用多麦克风语音增强技术和非封闭式前馈主动噪声控制技术用来提高噪声环境下通信耳机送、受话端语音的信噪比，保证通信语音的清晰度和可懂度。
其中，本发明在送话端提出结合用户佩戴状态识别的多麦克降噪技术，此技术不需要利用麦克风指向性来检测语音，而是通过检测麦克风所拾取的声音信号中主信号和参考信号的能量差异来识别不同的用户佩戴状态来相对应采用不同的降噪方法，从而保证在麦克风位置或形状不固定时降噪处理不会损伤语音。在受话端，本发明采用非封闭式前馈主动噪声控制技术，保证佩戴舒适性的同时有效降低语音频带内的噪声信号。
以下将结合附图对本发明的具体实施例进行详细描述。
本发明所提供的通信耳机的语音增强方法从送话端和受话端两方面进行降噪处理，由于本发明是在复用麦克风所采集到的声音信号的基础上进行降噪处理，所以本发明所应用的通信耳机中包括由至少两个麦克风构成的送话端、由至少一个麦克风和一个扬声器构成的受话端以及对声音信号进行降噪处理的主机。图2为应用本发明的一个通信耳机实施例的结构示意图。
如图2所示，本实施例所应用的通信耳机的入耳部分采用传统的非封闭入耳式耳塞结构，能够与耳朵较好耦合，提供佩戴的牢固性，又不会完全密封住耳道，保证长期佩戴的舒适性。通信耳机包括送话端、受话端、耳机线和主机230，其中，送话端利用三个麦克风所采集到的信号，麦克风212设置在耳机线上，麦克风214和216分别设置在耳机支杆的背部，开孔朝外。受话端包括两个麦克风214和216及两个扬声器224和226。
在这种通信耳机中，在耳机正常佩戴通话时使用者会把安装在耳机线上的麦克风212拿到嘴边（如图8所示），由于此麦克风212与嘴部距离较近，能够拾取信噪比比较高的声音信号，因此将该麦克风212视为主麦克风。而麦克风214和216由于设置在耳机支杆的背部，开孔朝外，在通信耳机的正常使用过程中距离使用者的嘴部距离较远，便于拾取较好的噪声参考信号，因此将这两个麦克风视为参考麦克风。
在本发明的一个具体实施方式中，采用三麦克风通信耳机300的系统方框图示于图3，其中主机端包括DSP单元200和模拟电路构成的受话端语音降噪处理单元700，DSP部分的送话端降噪处理单元400完成送话端语音的增强，同时啸叫检测单元500为受话端语音增强模块提供啸叫检测控制信号；受话端语音降噪处理单元700完成受话端的语音降噪。其中主机端可以采用DSP加一些模拟电路来单独实现，也可以作为一些音频设备或手机的一部分来实现。
需要说明的是，图3所示的实施例中所采用的麦克风的个数为3，但在本发明的具体应用中也可以采用其他数量的麦克风，如仅采用均设置在耳机支杆背部的两个麦克风（如214和216），此时送话端不存在主麦克风和参考麦克风之分，只使用单通道降噪模式即可；如果采用分设在耳机线和耳机支杆背部的两个麦克风（如212和214），则可以根据使用者的佩戴情况选择多麦克风降噪模式和/或单通道降噪模式；或者根据具体通信产品的需要采用更多的麦克风以达到更好的拾取有用语音信号和噪声信号的目的，此时可以根据麦克风所具体拾取到的声音信号判断是否有主、从麦克风的区分，从而采用相应的降噪模式。
下面将分别从送话端和受话端两大部分对本发明的语音增强方法和装置进行阐述。
图4为根据本发明的通信耳机的语音增强方法中送话端降噪处理部分的流程图。
如图4所示，送话端降噪处理部分的降噪处理流程包括：
S410：通过比较通信耳机受话端的麦克风拾取的声音信号的能量确定麦克风拾取信号能量差异，其中的声音信号包括语音信号和噪声信号；
S420：判断所确定的能量差异是否大于第一预置阈值来识别耳机佩戴状态，如果该能量差异大于第一预置阈值耳机正常佩戴（如图8所示），则进入步骤S430，否则，耳机非正常佩戴（如图9所示），进入步骤S440；
S430：对所拾取的声音信号进行多麦克风降噪；
S440：通过单通道降噪抑制声音信号中的稳态噪声。
图5为根据本发明实施例的在通信耳机送话端采用声信号处理方法进行语音增强的送话端降噪处理单元的逻辑结构示意图。
如图5所示，送话端降噪处理单元400包括佩戴状态确定模块420、多麦克风降噪模块440以及单通道降噪模块460。
其中，佩戴状态确定单元420用于通过比较构成送话端的麦克风拾取的声音信号的能量差异确定通信耳机的佩戴状态，如果能量差异大于第一预置阈值，确定所述通信耳机正常佩戴，否则，确定所述通信耳机非正常佩戴。其中，所拾取的声音信号包括语音信号和噪声信号；
多麦克风降噪模块440用于在上述能量差异大于第一预置阈值、通信耳机正常佩戴时，对麦克风所拾取的声音信号进行多麦克风降噪处理；
单通道降噪模块460用于在多麦克风降噪模块440对声音信号进行降噪处理后，进一步抑制残留的稳态噪声，以及，在上述能量差异小于等于第一预置阈值、通信耳机处于非正常佩戴状态时，直接对声音信号中的稳态噪声进行抑制处理。
下面结合图3、图4和图5对本发明在送话端的降噪处理方法以及降噪处理模块做更为细致的说明。
当通信耳机的耳塞处于佩戴状态时，麦克风214、216相对于人嘴的距离和位置都基本上是确定的，在本发明中被视为参考麦克风，它们所拾取的声音信号被视为参考信号。在正常应用状态中，麦克风212被拿到相对于人嘴的距离很近的位置，在本发明中被视为主麦克风，所拾取的声音信号被视为主信号。
但是麦克风212的位置在实际使用过程中却存在很大的不确定性。它可能与人嘴的距离比较近，也可能与麦克风214、216到人嘴的距离相当。通常定义麦克风212与人嘴距离较近时为正常佩戴模式，此时麦克风212拾取的主信号比麦克风214、216拾取的参考信号高，在通常的通信环境下处在送话状态时，一般主信号会比参考信号高6dB以上；而当麦克风212偏离人嘴过远时则为非正常佩戴模式，此时麦克风212 拾取到的主信号能量与麦克风214、216拾取的参考信号能量接近。通过这一特点，即可以在区分了主麦克风和参考麦克风的基础上，通过比较通信耳机的主、参考麦克风分别拾取的声音信号的能量差异来判断出耳机是否处于正常佩戴状态。
具体地，作为示例，在确定能量差异的过程中，首先把主麦克风212和参考麦克风214采集到的信号分别分为每N（N=512）个采样点的一帧数据，求这两帧数据的能量和P_112，P_114；然后求能量和的比值Rp=P_112/P_114。当Rp大于阈值Rth（如：Rth>6dB）时，为正常佩戴模式，此时利用多麦克风降噪单元460对声音信号进行多麦克降噪处理，然后再进行单通道降噪。当Rp小于阈值Rth，这时为非正常佩戴模式，不能很好区分语音与噪声，如果也同样采用多麦克降噪处理语音可能会被当成噪声抑制，因此只采用单通道降噪单元480作降噪处理，以避免语音损伤。
其中，多麦克风降噪模块440包括声音信号成分区分模块442和噪声信号衰减模块444。声音信号成分区分模块442用于通过比较声音信号中各频率成分的能量差异来区分声音信号中的语音信号成分和噪声信号成分；噪声信号衰减模块444用于对声音信号成分区分模块442所区分出的噪声信号成分进行衰减处理。
具体地，作为示例，当用户正常佩戴时，麦克风212拾取到的近场的语音信号成分要比麦克风214和216大6dB以上，而麦克风214、216和212拾取到的噪声成分能量相当。因此多麦克风降噪单元460利用麦克风212和麦克风214（也就是主麦克风和参考麦克风）拾取到的信号中各个频率成分的能量差异来区分语音成分和噪声成分，并对噪声成分进行降噪处理。
首先，利用声音信号成分区分模块442区分出语音信号和噪声信号。具体的处理过程包括：
把麦克风112和214的一帧数据分别进行快速傅里叶变换，把时域数据变换成各个频率成分Fi_112,Fi_114(i表示第i个频率成分)；
计算每个频率的能量Pi_112，Pi_114；对每个频率成分的能量进行比较得到能量比值Ri= Pi_112/Pi_114；
当Ri大于阈值Rthi（Rthi>6dB），则第i个频率成分是语音；当Ri小于阈值Rthi（Rthi>6dB），则第i个频率成分是噪声。
然后，保持语音成分，利用噪声信号衰减模块444衰减噪声成分。也就是说，当Ri大于阈值Rthi（Rthi>6dB）时，对于Fi_112不做处理；当Ri小于阈值Rthi（Rthi>6dB），对Fi_112乘上一个增益Gi（0<Gi<1），从而达到降噪的效果。
最后，把经过处理的Fi_112进行反傅里叶变换，就得到了经过降噪处理后的纯净语音信号。
本发明中的单通道降噪模块460的降噪原理是根据噪声平稳的特性统计出输入信号各个频带中的平稳噪声的能量进而消除。在本发明的一个实施方式中，单通道降噪模块460包括噪声能量统计模块462和噪声能量去除模块464，其中，噪声能量统计模块462用于利用平滑平均的方法统计出声音信号中各频率的噪声能量；噪声能量去除模块464用于在声音信号中去除噪声能量统计模块462所统计出的噪声能量，从而进一步降低了噪声成分并保留了语音成分，以达到提高语音信号信噪比的效果。
本发明在受话端采用前馈主动噪声控制方法进行降噪处理。通信耳机的入耳部分采用非封闭式的耳塞结构，主要是保证佩戴上耳机后耳道内气压是一致的，保证长期佩戴的舒适性。而前馈主动噪声控制麦克风一般置于通信耳机的外表面，尽可能多的拾取外界噪声的信息。因此应用在通信耳机上的这种前馈主动噪声控制的结构一般能满足系统对因果性的要求，从麦克风正面传过来的声音肯定是先到达麦克风处，然后再到达人耳处，从其他方向传过来的噪声因为要经过人头部的绕射所以基本也是先被麦克风拾取到。
图6为根据本发明的通信耳机的语音增强方法中受话端降噪处理部分的流程图。
如图6所示，本发明在受话端采用前馈主动噪声控制方法降低所接收的语音频带内的噪声信号的过程具体包括：
S610：利用通信耳机受话端的麦克风拾取噪声信号；
S620：根据所拾取的噪声信号确定反噪声信号；
S630：将所确定的反噪声信号与受话端接收的语音信号混合后通过构成受话端的扬声器馈入人耳，所述反噪声和入耳的原始噪声相互抵消而语音信号不变，从而降低所接收的语音频带内的噪声信号。
进而，在步骤S620根据噪声信号确定反噪声信号的过程中，首先利用反相器对噪声信号进行反相处理，确定初级反噪声信号；然后利用相位补偿器在音频范围内对初级反噪声信号的相位进行修正和调整，确定于所述噪声信号的相位完全反相的反噪声信号，并且采用包括双T型网络实现的有源滤波器，用来补偿低频部分由于非封闭式结构导致的低频相位损失。
图7为根据本发明实施例的受话端降噪处理单元的逻辑结构示意图。
如图7所示，受话端降噪处理单元700包括噪声信号确定模块720、反噪声信号确定模块740和输出信号混合模块760，其中，反噪声信号确定模块740可以包括反相器743和相位补偿器744。
噪声信号拾取模块720用于利用通信耳机受话端的麦克风拾取噪声信号，由于受话端在接收远端语音信号时，麦克风所拾取的声音信号一般全部视为噪声信号，因此，设置在耳机支杆的背部的麦克风214和216也就相当于此处的噪声信号拾取模块720。反噪声信号确定模块740用于根据噪声信号确定模块720所确定的噪声信号确定反噪声信号；输出信号混合模块760用于将反噪声信号确定模块740所确定的反噪声信号与所述受话端接收的语音信号混合后通过构成受话端的扬声器224馈入人耳，该反噪声和入耳的原始噪声（由自然的声学通道传递）相互抵消而语音信号不变，从而降低所接收的语音频带内的噪声信号。
反相器742用于对所述噪声信号进行反相处理，确定初级反噪声信号；
相位补偿器744用于在音频范围内对初级反噪声信号的相位进行修正和调整，确定与噪声信号的相位完全反相的反噪声信号，并且采用包括双T型网络实现的有源滤波器，用来补偿低频部分由于非封闭式结构导致的低频相位损失。
另外，受话端降噪处理单元700还可以包括第一放大器730和第二放大器750，第一放大器730用以对噪声信号拾取模块720所拾取的噪声信号进行放大，第二放大器750用于对混有反噪声信号和语音信号的混合信号进行放大。
具体地，作为示例，麦克风214拾取到的噪声信号经过前置第一放大器730放大，然后经过一个反相器742和相位补偿器744，来产生与原始噪声幅度相同相位相反的反噪声信号。
相位补偿器744主要是为了解决前馈主动噪声控制技术应用于非封闭式通信耳机所存在的时延问题，从电路上在音频范围内对反噪声信号的相位进行相应的修正和调整，以期达到反噪声和原始噪声相位刚好反相的目的。其一般的实现方式是采用无源或有源的双T型网络来实现。
反噪声信号和输入的语音信号经过一个由加法器构成的输出信号混合模块760在电路上混合，作为后端第二放大器750的输入，第二放大器750把混有反噪声和语音信号的混合信号放大后直接去推动扬声器224。
同样，麦克风216拾取到的噪声信号经过前置第一放大器730放大、反相器742反相、相位补偿器744补偿以及加法器的混合、第二放大器750的放大后，直接去推动扬声器226。
麦克风前置第一放大器730、反相器742、相位补偿器744、加法器、扬声器功率第二放大器750等部分可以采用单独的器件实现也可以由同一器件来实现一个或者几个模块的功能。
混有反噪声和语音的混合信号经过扬声器，变成声信号馈入人耳，从扬声器发出的反噪声信号和从声学通道传播到人耳处的原始噪声信号幅度相同相位相反，所以会在人耳处发生相互叠加抵消，从而达到同时消除原始噪声和反噪声的目的，噪声得到了降低，而语音能量不变，有效提高了语音信号的信噪比，传播到人耳处的将是清晰可懂较纯净的语音信号。
对于传统的采用封闭式的前馈主动噪声控制耳机，外界噪声从参考麦克风传播到人耳处需要经过被动隔声材料，会增加声学通道的延迟，从而为电子通道提供更多的处理时间，保证了系统的因果性。为了解决前馈主动噪声控制技术应用于非封闭式结构通信耳机的时延问题，需要从二个方面对系统进行很好的选择和设计，首先需要对扬声器单体的前、后腔进行较好设计和处理，调整前后腔的大小和开孔以改善由扬声器到人耳之间在音频范围内的相位响应，其次需要在电路上对反相器进行相位补偿，从电路本身对时延进行修正和补偿，以期在整个音频范围内都有较好的降噪效果。
从麦克风和人耳处的距离设计来看，一方面希望越近越好，麦克风与人耳靠得越近，两点的噪声相关性就越好，噪声消除就能做的越好；但另一方面要求麦克风和人耳有一定的距离，以保证当噪声从麦克风处传播到人耳这段时间内，电子层面上有较多的处理时间。另外，麦克风与扬声器之间也需要保持一定的空间距离并较好的进行声学隔离，避免由扬声器发出的信号被麦克风拾取到，从而避免麦克风拾取的噪声信号里面包括有用的语音信号和导致整个系统形成反馈啸叫的反馈回路，如果存在反馈回路，系统增益过高的话就有可能出现啸叫现象。
另外，对于非封闭式前馈主动降噪耳机，扬声器和拾取外界噪声的参考麦克风之间存在固有的泄露通道。在耳机正常佩戴时，从扬声器到参考麦克风之间的声学传递函数幅度非常小，所以正常使用时，非封闭式前馈主动噪声控制技术不会对语音信号造成损伤，系统也不会存在不稳定的啸叫现象。但当把耳机放在一个密闭或者半密闭空间内时，扬声器到参考麦克风之间的声学传递函数的幅度将急剧增大，特别是高频部分。
这种较大幅度的声学传递函数，与具有高增益的控制电路，就构成了一个闭环反馈系统，当闭环反馈系统的幅度和相位满足一定条件时，系统将发生自激啸叫，存在鲁棒性问题。
因此，在本发明的一个优选实施方式中，DSP单元还包括啸叫检测模单元，用于为受话端语音增强模块提供啸叫检测控制信号，具体而言，就是在通信耳机的麦克风所拾取的声音信号的频谱中某一频率点的能量相比于其他频带的能量高出预设值、并且该频率点的能量不断增加时，通过控制信号自主调整对受话端的降噪处理。
通常，如果判断出在某个频率点的能量比旁边其他频带的能量高10dB以上并在此频率能量不断增加，将判断系统处在不正常状态，啸叫检测单元将输出控制信号，调整主动噪声控制电路。控制方式可以采用减小第一放大器的增益或者直接切断主动噪声控制电路电源来实现。
对于送话端、受话端的语音信号，可以采用有线的方式和其他设备连接，也可以采用蓝牙等无线连接方式和其他设备连接。
以上结合附图和多个具体实施方式对本发明的用于提高噪声环境中通信耳机送、受话端语音信噪比的技术和装置。显然，本领域的普通技术人员在不偏离本发明概念的条件下，不需要付出创造性的劳动，就可以对在此公开的特定装置和技术进行许多应用和修改并且可以与在此所公开的特定装置和技术有所不同。因此，本发明应该被理解成包括此处所公开的装置和技术所提出或拥有的每一个新颖特征和新颖的特征组合，但凡本领域普通技术人员根据本发明所揭示的内容所作的等效修饰和变化，皆应纳入权利要求书中记载的保护范围内。

Claims (13)

  1. 一种通信耳机的语音增强方法，所述通信耳机包括由至少两个麦克风构成的送话端和由至少一个麦克风及一个扬声器构成的受话端，所述方法复用多个麦克风信号在所述通信耳机的送话端和受话端分别提供降噪处理，其中，在所述送话端的降噪处理包括：
    通过比较构成送话端的麦克风拾取的声音信号的能量差异确定所述通信耳机的佩戴状态；
    如果所述能量差异大于第一预置阈值，确定所述通信耳机正常佩戴，则先对所述声音信号进行多麦克风降噪，然后通过单通道降噪进一步抑制残留的稳态噪声；否则，
    确定所述通信耳机非正常佩戴，直接通过单通道降噪抑制所述声音信号中的稳态噪声。
  2. 如权利要求1所述的通信耳机的语音增强方法，其中，在对所述声音信号进行多麦克风降噪的过程中，具体包括：
    通过比较所述声音信号中各频率成分的能量差异来区分所述声音信号中的语音信号成分和噪声信号成分；
    对所述噪声信号成分进行衰减处理。
  3. 如权利要求2所述的通信耳机的语音增强方法，其中，在通过比较所述声音信号中各频率成分的能量差异来区分所述声音信号中的语音信号成分和噪声信号成分的过程中，
    如果所述声音信号中某频率成分的能量差异大于第二预置阈值，则将能量差异大于所述第二预置阈值的该频率成分作为语音信号成分；
    如果所述声音信号中某频率成分的能量差异小于等于所述第二预置阈值，则将能量差异小于等于所述第二预置阈值的该频率成分作为噪声信号成分。
  4. 如权利要求1所述的通信耳机的语音增强方法，其中，在通过单通道降噪抑制稳态噪声的过程中，具体包括：
    利用平滑平均的方法统计出所述声音信号中各频率的噪声能量；
    在所述声音信号中去除所述噪声能量。
  5. 如权利要求1～4中任一项所述的通信耳机的语音增强方法，其中，所述通信耳机的入耳部分采用非封闭式耳塞结构，正常佩戴状态下所述通信耳机的扬声器与人耳耳道之间的耦合位置相对固定；并且，在所述受话端降噪处理包括：
    利用构成所述受话端的麦克风拾取噪声信号；
    根据所述噪声信号确定反噪声信号；
    将所述反噪声信号与所述受话端接收的语音信号混合后通过构成受话端的扬声器馈入人耳。
  6. 如权利要求5所述的通信耳机的语音增强方法，其中，在根据所述噪声信号确定反噪声信号的过程中，
    首先利用反相器对所述噪声信号进行反相处理，确定初级反噪声信号；
    然后利用相位补偿器在音频范围内对所述初级反噪声信号的相位进行修正和调整，确定与所述噪声信号的相位完全反相的反噪声信号，其中所述相位补偿器包括采用双T型网络实现的有源滤波器，用来补偿低频部分由于非封闭式结构导致的低频相位损失。
  7. 如权利要求1所述的通信耳机的语音增强方法，其中，所述方法还包括检测并抑制啸叫的处理，具体包括：
    如果在所述通信耳机的麦克风拾取的声音信号的频谱中某一频率点的能量比其他频带的能量高出预设值，并且所述频率点的能量不断增加，则自主调整对所述受话端的降噪处理。
  8. 一种通信耳机，所述通信耳机包括由至少两个麦克风构成的送话端和由至少一个麦克风和一个扬声器构成的受话端，以及送话端降噪处理单元和受话端降噪处理单元，其中，所述送话端降噪处理单元包括：
    佩戴状态确定模块，用于通过比较构成所述送话端的麦克风拾取的声音信号的能量差异确定所述通信耳机的佩戴状态，如果所述能量差异大于第一预置阈值，确定所述通信耳机正常佩戴，否则，确定所述通信耳机非正常佩戴；
    多麦克风降噪模块，用于在所述通信耳机正常佩戴时，对所述声音信号进行多麦克风降噪处理；
    单通道降噪模块，用于在所述多麦克风降噪模块对所述声音信号进行降噪处理后，进一步抑制残留的稳态噪声，以及，在所述通信耳机非正常佩戴时，直接对所述声音信号中的稳态噪声进行抑制处理。
  9. 如权利要求8所述的通信耳机，其中，所述多麦克风降噪模块进一步包括：
    声音信号成分区分模块，用于通过比较所述声音信号中各频率成分的能量差异来区分所述声音信号中的语音信号成分和噪声信号成分；
    噪声信号衰减模块，用于对所述噪声信号成分进行衰减处理。
  10. 如权利要求8所述的通信耳机，其中，所述单通道降噪模块进一步包括：
    噪声能量统计模块，用于利用平滑平均的方法统计出所述声音信号中各频率的噪声能量；
    噪声能量去除模块，用于在所述声音信号中去除所述噪声能量。
  11. 如权利要求8～10中任一项所述的通信耳机，其中，所述通信耳机的入耳部分采用非封闭式耳塞结构，正常佩戴状态下所述通信耳机的扬声器与人耳耳道之间的耦合位置相对固定；并且，所述受话端降噪处理单元包括：
    噪声信号拾取模块，用于利用构成所述受话端的麦克风拾取噪声信号；
    反噪声信号确定模块，用于根据所述噪声信号确定反噪声信号；
    输出信号混合模块，用于将所述反噪声信号与所述受话端接收的语音信号混合后通过构成受话端的扬声器馈入人耳。
  12. 如权利要求11所述的通信耳机，其中，所述反噪声信号确定模块进一步包括：
    反相器，用于对所述噪声信号进行反相处理，确定初级反噪声信号；
    相位补偿器，用于在音频范围内对所述初级反噪声信号的相位进行修正和调整，确定与所述噪声信号的相位完全反相的反噪声信号，并且采用包括双T型网络实现的有源滤波器，用来补偿低频部分由于非封闭式结构导致的低频相位损失。
  13. 如权利要求8所述的通信耳机，所述通信耳机进一步包括：
    啸叫检测单元，用于在所述通信耳机的麦克风所拾取的声音信号的频谱中某一频率点的能量相比于其他频带的能量高出预设值、并且所述频率点的能量不断增加时，通过控制信号自主调整对所述受话端的降噪处理。
PCT/CN2012/072483 2011-08-10 2012-03-16 一种通信耳机的语音增强方法、装置及降噪通信耳机 WO2013020380A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2013552095A JP5513690B2 (ja) 2011-08-10 2012-03-16 通信イヤホンの音声増強方法、装置及びノイズ低減通信イヤホン
US14/110,879 US9484042B2 (en) 2011-08-10 2012-03-16 Speech enhancing method, device for communication earphone and noise reducing communication earphone
EP12822487.0A EP2680608B1 (en) 2011-08-10 2012-03-16 Communication headset speech enhancement method and device, and noise reduction communication headset
KR1020137021521A KR101353686B1 (ko) 2011-08-10 2012-03-16 통신 이어폰의 음성보강 방법, 장치 및 노이즈저감 통신 이어폰
DK12822487.0T DK2680608T3 (en) 2011-08-10 2012-03-16 Speech Improvement Process and decoration for communication headsets and communications headset with noise reduction

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2011102290039A CN102300140B (zh) 2011-08-10 2011-08-10 一种通信耳机的语音增强方法及降噪通信耳机
CN201110229003.9 2011-08-10

Publications (1)

Publication Number Publication Date
WO2013020380A1 true WO2013020380A1 (zh) 2013-02-14

Family

ID=45360247

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/072483 WO2013020380A1 (zh) 2011-08-10 2012-03-16 一种通信耳机的语音增强方法、装置及降噪通信耳机

Country Status (7)

Country Link
US (1) US9484042B2 (zh)
EP (1) EP2680608B1 (zh)
JP (1) JP5513690B2 (zh)
KR (1) KR101353686B1 (zh)
CN (1) CN102300140B (zh)
DK (1) DK2680608T3 (zh)
WO (1) WO2013020380A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106792315A (zh) * 2017-01-05 2017-05-31 歌尔科技有限公司 一种抵消环境噪声的方法和装置及一种主动降噪耳机
CN107750028A (zh) * 2017-12-06 2018-03-02 贵州翔通科技实业有限公司 耳机
CN111696564A (zh) * 2020-06-05 2020-09-22 北京搜狗科技发展有限公司 语音处理方法、装置和介质
CN112822592A (zh) * 2020-12-31 2021-05-18 青岛理工大学 一种可定向聆听的有源降噪耳机及控制方法
CN114007157A (zh) * 2021-10-28 2022-02-01 中北大学 一种智能降噪通信耳机

Families Citing this family (104)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102300140B (zh) 2011-08-10 2013-12-18 歌尔声学股份有限公司 一种通信耳机的语音增强方法及降噪通信耳机
CN102543097A (zh) * 2012-01-16 2012-07-04 华为终端有限公司 降噪方法及设备
CN102625207B (zh) * 2012-03-19 2015-09-30 中国人民解放军总后勤部军需装备研究所 一种主动式噪声防护耳塞的声音信号处理方法
CN102710839B (zh) * 2012-04-27 2017-11-28 华为技术有限公司 一种提升语音通话效果的方法及通信终端
CN102801861B (zh) * 2012-08-07 2015-08-19 歌尔声学股份有限公司 一种应用于手机的语音增强方法和装置
CN103680511B (zh) * 2012-09-24 2017-03-01 联想(北京)有限公司 一种滤除噪音的方法、装置和电子设备
CN103873980B (zh) * 2012-12-10 2017-11-28 联想(北京)有限公司 音频处理方法、装置及电子设备
JP2014155144A (ja) * 2013-02-13 2014-08-25 Funai Electric Co Ltd 音声入力装置及び雑音抑圧方法
CN104335600B (zh) * 2013-02-25 2017-08-08 展讯通信（上海）有限公司 多麦克风移动装置中检测及切换降噪模式的方法
CN103269465B (zh) * 2013-05-22 2016-09-07 歌尔股份有限公司 一种强噪声环境下的耳机通讯方法和一种耳机
US9264803B1 (en) 2013-06-05 2016-02-16 Google Inc. Using sounds for determining a worn state of a wearable computing device
CN103338419B (zh) * 2013-06-29 2016-07-06 青岛歌尔声学科技有限公司 一种消除耳机啸叫的方法及装置
CN103391496B (zh) * 2013-07-16 2016-08-10 歌尔声学股份有限公司 应用于主动噪声消除anr耳机的啸叫抑制方法和装置
CN104424953B (zh) * 2013-09-11 2019-11-01 华为技术有限公司 语音信号处理方法与装置
CN103632009B (zh) * 2013-12-13 2016-08-17 南京大学 一种模拟反馈有源降噪耳机的设计方法
CN104754430A (zh) * 2013-12-30 2015-07-01 重庆重邮信科通信技术有限公司 终端麦克风降噪装置和方法
CN105025418B (zh) * 2014-04-17 2018-10-09 山东共达电声股份有限公司 一种主动降噪控制器
KR101609194B1 (ko) 2014-07-21 2016-04-05 대진반도체 주식회사 4극 오디오 플러그-잭 커넥터 방식 액세서리 장치 및 전원공급방법과 호스트 장치
JP6204312B2 (ja) * 2014-08-28 2017-09-27 日本電信電話株式会社 収音装置
KR20160033490A (ko) 2014-09-18 2016-03-28 박태준 4극 오디오 플러그-잭 커넥터 방식 액세서리 장치 및 전원공급방법
US9786261B2 (en) * 2014-12-15 2017-10-10 Honeywell International Inc. Active noise reduction earcup with speaker array
JP2016127502A (ja) * 2015-01-06 2016-07-11 富士通株式会社 通信装置及びプログラム
CN104602157B (zh) * 2015-01-28 2018-11-02 深圳市冠旭电子股份有限公司 一种耳机降噪方法及装置
CN104754436B (zh) * 2015-03-13 2024-01-16 钰太芯微电子科技（上海）有限公司 一种主动降噪方法以及降噪耳机
EP3278575B1 (en) * 2015-04-02 2021-06-02 Sivantos Pte. Ltd. Hearing apparatus
CN106210219B (zh) * 2015-05-06 2019-03-22 小米科技有限责任公司 降噪方法及装置
US9843859B2 (en) 2015-05-28 2017-12-12 Motorola Solutions, Inc. Method for preprocessing speech for digital audio quality improvement
CN105120403B (zh) * 2015-06-26 2018-08-17 努比亚技术有限公司 一种降噪系统和方法
CN106328116B (zh) * 2015-06-30 2020-04-17 芋头科技（杭州）有限公司 一种机器人室内噪声控制系统
KR101731714B1 (ko) 2015-08-13 2017-04-28 중소기업은행 음질 개선을 위한 방법 및 헤드셋
US9401158B1 (en) * 2015-09-14 2016-07-26 Knowles Electronics, Llc Microphone signal fusion
CN110493692B (zh) 2015-10-13 2022-01-25 索尼公司 信息处理装置
CN105163238A (zh) * 2015-10-17 2015-12-16 深圳跃豁达科技有限公司 一种主动深度降噪耳机
CN105472137A (zh) * 2015-11-19 2016-04-06 广东小天才科技有限公司 一种调整通话音量的方法及装置
CN105451111B (zh) * 2015-12-10 2019-03-19 小米科技有限责任公司 耳机播放控制方法、装置及终端
CN105511641A (zh) * 2016-01-09 2016-04-20 温州智之粹知识产权有限公司 一种声控键盘
CN105744429A (zh) * 2016-05-17 2016-07-06 乐视控股（北京）有限公司 基于移动终端的耳机降噪方法、移动终端及降噪耳机
US11030988B2 (en) * 2016-09-30 2021-06-08 Sony Corporation Signal processing device, signal processing method, and program
US10354639B2 (en) * 2016-10-24 2019-07-16 Avnera Corporation Automatic noise cancellation using multiple microphones
CN106534461B (zh) * 2016-11-04 2019-07-26 惠州Tcl移动通信有限公司 耳机的降噪系统及其降噪方法
KR102535726B1 (ko) * 2016-11-30 2023-05-24 삼성전자주식회사 이어폰 오장착 검출 방법, 이를 위한 전자 장치 및 저장 매체
CN106791141A (zh) * 2016-12-30 2017-05-31 努比亚技术有限公司 一种通话音效参数的调整方法及移动终端
TWI627583B (zh) * 2017-01-04 2018-06-21 元鼎音訊股份有限公司 具有通知功能之發聲裝置、資料分析通知系統及其資料分析之方法
US10237654B1 (en) * 2017-02-09 2019-03-19 Hm Electronics, Inc. Spatial low-crosstalk headset
KR101898911B1 (ko) * 2017-02-13 2018-10-31 주식회사 오르페오사운드웍스 인이어 마이크와 아웃이어 마이크 수음특성을 이용한 소음 제거 이어셋 및 소음 제거 방법
CN107071608B (zh) * 2017-02-14 2023-09-29 歌尔股份有限公司 降噪耳机以及电子设备
CN106981294A (zh) * 2017-02-17 2017-07-25 成都易慧通科技有限公司 一种基于亚音频特征判断的啸叫消除方法
US9894452B1 (en) 2017-02-24 2018-02-13 Bose Corporation Off-head detection of in-ear headset
EP3593349B1 (en) * 2017-03-10 2021-11-24 James Jordan Rosenberg System and method for relative enhancement of vocal utterances in an acoustically cluttered environment
US10311889B2 (en) * 2017-03-20 2019-06-04 Bose Corporation Audio signal processing for noise reduction
US10366708B2 (en) 2017-03-20 2019-07-30 Bose Corporation Systems and methods of detecting speech activity of headphone user
CN106952654A (zh) * 2017-04-24 2017-07-14 北京奇虎科技有限公司 机器人降噪方法、装置及机器人
US10468020B2 (en) * 2017-06-06 2019-11-05 Cypress Semiconductor Corporation Systems and methods for removing interference for audio pattern recognition
CN108093327B (zh) * 2017-09-15 2019-11-29 歌尔科技有限公司 一种检验耳机佩戴一致性的方法、装置和电子设备
CN109729463A (zh) * 2017-10-27 2019-05-07 北京金锐德路科技有限公司 用于脖戴式语音交互耳机的声麦骨麦复合收音装置
CN109729457A (zh) * 2017-10-27 2019-05-07 北京金锐德路科技有限公司 用于脖戴式语音交互耳机的骨麦收音处理装置
CN107889002B (zh) * 2017-10-30 2019-08-27 恒玄科技（上海）有限公司 颈环蓝牙耳机、颈环蓝牙耳机的降噪系统及降噪方法
EP3725093B1 (en) 2017-12-15 2023-01-18 GN Audio A/S A headset with ambient noise reduction system
EP3503573A1 (en) * 2017-12-20 2019-06-26 GN Hearing A/S Hearing protection device with reliability and related methods
CN108198565B (zh) * 2017-12-28 2020-11-17 深圳市东微智能科技股份有限公司 混音处理方法、装置、计算机设备和存储介质
CN108305637B (zh) * 2018-01-23 2021-04-06 Oppo广东移动通信有限公司 耳机语音处理方法、终端设备及存储介质
CN108063996A (zh) * 2018-01-24 2018-05-22 江西联创宏声万安电子有限公司 一种Type C接口耳机及其实现方法
CN108174321B (zh) * 2018-02-11 2019-09-17 歌尔股份有限公司 耳机降噪处理的方法、装置及耳机
CN108449687B (zh) * 2018-03-13 2019-04-26 江苏华腾智能科技有限公司 一种多麦克风阵列降噪的会议系统
US10438605B1 (en) * 2018-03-19 2019-10-08 Bose Corporation Echo control in binaural adaptive noise cancellation systems in headsets
TWI672690B (zh) * 2018-03-21 2019-09-21 塞席爾商元鼎音訊股份有限公司 人工智慧語音互動之方法、電腦程式產品及其近端電子裝置
DE102019107173A1 (de) * 2018-03-22 2019-09-26 Sennheiser Electronic Gmbh & Co. Kg Verfahren und Vorrichtung zum Erzeugen und Ausgeben eines Audiosignals zum Erweitern des Höreindrucks bei Live-Veranstaltungen
CN108847208B (zh) * 2018-05-04 2020-11-27 歌尔科技有限公司 一种降噪处理方法、装置和耳机
KR101926817B1 (ko) * 2018-05-17 2019-03-07 주식회사 지에스씨 하울링 및 소음제거기능을 구비한 페이지 폰
US10885896B2 (en) * 2018-05-18 2021-01-05 Bose Corporation Real-time detection of feedforward instability
CN108540887B (zh) * 2018-05-21 2019-12-20 歌尔股份有限公司 一种头戴式降噪耳机和降噪处理方法
CN108810692A (zh) * 2018-05-25 2018-11-13 会听声学科技（北京）有限公司 主动降噪系统、主动降噪方法及耳机
CN108566684B (zh) * 2018-07-02 2021-04-13 Oppo广东移动通信有限公司 通信连接建立方法及相关设备
CN108847250B (zh) * 2018-07-11 2020-10-02 会听声学科技（北京）有限公司 一种定向降噪方法、系统及耳机
CN112384975A (zh) * 2018-07-12 2021-02-19 杜比实验室特许公司 使用辅助信号的音频装置的传输控制
CN110740396A (zh) * 2018-07-18 2020-01-31 安克创新科技股份有限公司 一种降噪耳机
CN109089201B (zh) * 2018-07-26 2020-04-17 Oppo广东移动通信有限公司 麦克风堵孔检测方法及相关产品
CN109218882B (zh) * 2018-08-16 2021-02-26 歌尔科技有限公司 耳机的环境声音监听方法及耳机
CN110891226B (zh) * 2018-09-07 2022-06-24 中兴通讯股份有限公司 一种消噪方法、装置、设备和存储介质
CN109246513A (zh) * 2018-09-30 2019-01-18 歌尔科技有限公司 一种主动降噪耳机及其降噪方法、装置
US10462551B1 (en) 2018-12-06 2019-10-29 Bose Corporation Wearable audio device with head on/off state detection
CN110677796B (zh) * 2019-03-14 2021-12-17 深圳攀高医疗电子有限公司 一种音频信号处理方法及助听器
CN110300344A (zh) * 2019-03-25 2019-10-01 深圳市增长点科技有限公司 自适应降噪耳机
CN110021307B (zh) * 2019-04-04 2022-02-01 Oppo广东移动通信有限公司 音频校验方法、装置、存储介质及电子设备
CN110012378B (zh) * 2019-04-08 2020-10-16 深圳市九音科技有限公司 一种语音降噪的方法、耳塞和计算机存储介质
CN110197670B (zh) * 2019-06-04 2022-06-07 大众问问(北京)信息科技有限公司 音频降噪方法、装置及电子设备
US11109156B2 (en) 2019-07-24 2021-08-31 Barefoot Sound, Llc Method and apparatus for improving effective signal-to-noise ratio of analog to digital conversion for multi-band digital signal processing devices
CN110430500B (zh) * 2019-07-31 2020-08-04 歌尔科技有限公司 耳机的降噪方法和耳机
CN110853665B (zh) * 2019-11-28 2022-04-26 歌尔科技有限公司 一种数字麦克风降噪方法、系统、设备及计算机存储介质
CN111103807A (zh) * 2019-12-17 2020-05-05 青岛海信智慧家居系统股份有限公司 一种家用终端设备的控制方法及装置
CN111010642B (zh) * 2019-12-25 2022-06-10 歌尔股份有限公司 一种耳机及其上行降噪方法
CN113038318B (zh) * 2019-12-25 2022-06-07 荣耀终端有限公司 一种语音信号处理方法及装置
CN113973249B (zh) * 2020-07-24 2023-04-07 华为技术有限公司 耳机通话方法及耳机
CN112019977A (zh) * 2020-09-04 2020-12-01 广州郝舜科技有限公司 一种大数据采集用音频采集装置
CN112116918B (zh) * 2020-09-27 2023-09-22 北京声加科技有限公司 语音信号增强处理方法和耳机
CN112929800B (zh) * 2021-02-04 2022-08-12 歌尔科技有限公司 拾音装置、电子设备及拾音方法
CN113223508B (zh) * 2021-03-29 2023-08-04 深圳市芯中芯科技有限公司 一种双模tws蓝牙耳机的管理方法
CN113596662B (zh) * 2021-07-30 2024-04-02 北京小米移动软件有限公司 啸叫声的抑制方法、啸叫声的抑制装置、耳机及存储介质
CN113808441B (zh) * 2021-08-03 2023-07-07 郑州科技学院 便携式外语单词练习器
CN113938782B (zh) * 2021-09-30 2023-10-31 安克创新科技股份有限公司 耳机入耳状态的识别及耳机模式的自适应调节方法和耳机
CN114257684A (zh) * 2021-12-17 2022-03-29 歌尔科技有限公司 一种语音处理方法、系统、装置及电子设备
CN116614742A (zh) * 2023-07-20 2023-08-18 江西红声技术有限公司 一种清晰语音送受话降噪耳机
CN116801157A (zh) * 2023-08-28 2023-09-22 深圳市鑫正宇科技有限公司 一种无线耳机组件及其信号处理方法
CN116939428B (zh) * 2023-09-18 2023-12-22 歌尔股份有限公司 耳机设备、风噪抑制方法及计算机可读存储介质

Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1507293A (zh) 2002-12-09 2004-06-23 Ӣҵ�Ｏ�ţ��Ϻ������Ӽ������޹�˾ 智能化调节移动通讯装置耳机音量的方法
EP1489885A2 (de) * 2003-06-20 2004-12-22 Siemens Audiologische Technik GmbH Verfahren zum Betrieb eines Hörhilfegerätes sowie Hörhilfegerät mit einem Mikrofonsystem, bei dem unterschiedliche Richtcharakteristiken einstellbar sind
CN1684143A (zh) 2004-04-14 2005-10-19 华为技术有限公司 一种语音增强的方法
CN1967658A (zh) * 2005-11-14 2007-05-23 北京大学科技开发部 小尺度麦克风阵列语音增强系统和方法
CN1967158A (zh) 2006-08-18 2007-05-23 上海一诺仪表有限公司 插入式电旋流量计
WO2008056649A1 (fr) * 2006-11-09 2008-05-15 Panasonic Corporation Détecteur de position de source sonore
CN101432798A (zh) 2006-04-26 2009-05-13 伯斯有限公司 利用主动降噪的头戴式耳机的通信
CN101466055A (zh) 2008-12-31 2009-06-24 瑞声声学科技(常州)有限公司 小型麦克风阵列装置及其波束形成方法
CN101466056A (zh) 2008-12-31 2009-06-24 瑞声声学科技(常州)有限公司 麦克风消噪方法及装置
CN101477800A (zh) 2008-12-31 2009-07-08 瑞声声学科技(深圳)有限公司 语音增强的方法
CN101777349A (zh) * 2009-12-08 2010-07-14 中国科学院自动化研究所 基于听觉感知特性的信号子空间麦克风阵列语音增强方法
CN101790752A (zh) * 2007-09-28 2010-07-28 高通股份有限公司 多麦克风声音活动检测器
CN101853667A (zh) 2010-05-25 2010-10-06 无锡中星微电子有限公司 一种语音降噪装置
CN102074236A (zh) * 2010-11-29 2011-05-25 清华大学 一种分布式麦克风的说话人聚类方法
CN102300140A (zh) * 2011-08-10 2011-12-28 歌尔声学股份有限公司 一种通信耳机的语音增强方法、装置及降噪通信耳机

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5125032A (en) * 1988-12-02 1992-06-23 Erwin Meister Talk/listen headset
US5732143A (en) * 1992-10-29 1998-03-24 Andrea Electronics Corp. Noise cancellation apparatus
JP2000330597A (ja) * 1999-05-20 2000-11-30 Matsushita Electric Ind Co Ltd 雑音抑圧装置
US20030053650A1 (en) * 2001-09-20 2003-03-20 Kuo-Shou Wang Earphone device for motorcycle rider and passenger
JP2003241792A (ja) * 2002-02-22 2003-08-29 Matsushita Electric Works Ltd 音声認識装置及び方法
WO2003083828A1 (en) * 2002-03-27 2003-10-09 Aliphcom Nicrophone and voice activity detection (vad) configurations for use with communication systems
JP4058987B2 (ja) 2002-04-15 2008-03-12 三菱電機株式会社 雑音除去装置及び雑音除去方法
EP1581026B1 (en) * 2004-03-17 2015-11-11 Nuance Communications, Inc. Method for detecting and reducing noise from a microphone array
US8189803B2 (en) * 2004-06-15 2012-05-29 Bose Corporation Noise reduction headset
US20080076489A1 (en) * 2006-08-07 2008-03-27 Plantronics, Inc. Physically and electrically-separated, data-synchronized data sinks for wireless systems
US7773759B2 (en) * 2006-08-10 2010-08-10 Cambridge Silicon Radio, Ltd. Dual microphone noise reduction for headset application
WO2009076523A1 (en) * 2007-12-11 2009-06-18 Andrea Electronics Corporation Adaptive filtering in a sensor array system
KR200447028Y1 (ko) 2007-12-13 2009-12-21 주식회사 모닝테크놀로지 청각 장애인을 위한 종합 통신 시스템
US8818000B2 (en) * 2008-04-25 2014-08-26 Andrea Electronics Corporation System, device, and method utilizing an integrated stereo array microphone
GB2461315B (en) * 2008-06-27 2011-09-14 Wolfson Microelectronics Plc Noise cancellation system
US20100020998A1 (en) * 2008-07-28 2010-01-28 Plantronics, Inc. Headset wearing mode based operation
JP5417821B2 (ja) * 2008-11-28 2014-02-19 ソニー株式会社 音声信号再生装置、携帯電話端末
US8340312B2 (en) * 2009-08-04 2012-12-25 Apple Inc. Differential mode noise cancellation with active real-time control for microphone-speaker combinations used in two way audio communications
US8223986B2 (en) * 2009-11-19 2012-07-17 Apple Inc. Electronic device and external equipment with digital noise cancellation and digital audio path
CN101778333B (zh) * 2010-01-27 2013-01-16 杭州华三通信技术有限公司 一种麦克风状态的检测方法和装置
EP2555189B1 (en) * 2010-11-25 2016-10-12 Goertek Inc. Method and device for speech enhancement, and communication headphones with noise reduction
KR101364543B1 (ko) * 2011-11-17 2014-02-19 한양대학교 산학협력단 휴대 단말기를 이용한 음향 수신 장치 및 방법

Patent Citations (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1507293A (zh) 2002-12-09 2004-06-23 Ӣҵ�Ｏ�ţ��Ϻ������Ӽ������޹�˾ 智能化调节移动通讯装置耳机音量的方法
EP1489885A2 (de) * 2003-06-20 2004-12-22 Siemens Audiologische Technik GmbH Verfahren zum Betrieb eines Hörhilfegerätes sowie Hörhilfegerät mit einem Mikrofonsystem, bei dem unterschiedliche Richtcharakteristiken einstellbar sind
CN1684143A (zh) 2004-04-14 2005-10-19 华为技术有限公司 一种语音增强的方法
CN1967658A (zh) * 2005-11-14 2007-05-23 北京大学科技开发部 小尺度麦克风阵列语音增强系统和方法
CN101432798A (zh) 2006-04-26 2009-05-13 伯斯有限公司 利用主动降噪的头戴式耳机的通信
CN1967158A (zh) 2006-08-18 2007-05-23 上海一诺仪表有限公司 插入式电旋流量计
WO2008056649A1 (fr) * 2006-11-09 2008-05-15 Panasonic Corporation Détecteur de position de source sonore
CN101790752A (zh) * 2007-09-28 2010-07-28 高通股份有限公司 多麦克风声音活动检测器
CN101466055A (zh) 2008-12-31 2009-06-24 瑞声声学科技(常州)有限公司 小型麦克风阵列装置及其波束形成方法
CN101466056A (zh) 2008-12-31 2009-06-24 瑞声声学科技(常州)有限公司 麦克风消噪方法及装置
CN101477800A (zh) 2008-12-31 2009-07-08 瑞声声学科技(深圳)有限公司 语音增强的方法
CN101777349A (zh) * 2009-12-08 2010-07-14 中国科学院自动化研究所 基于听觉感知特性的信号子空间麦克风阵列语音增强方法
CN101853667A (zh) 2010-05-25 2010-10-06 无锡中星微电子有限公司 一种语音降噪装置
CN102074236A (zh) * 2010-11-29 2011-05-25 清华大学 一种分布式麦克风的说话人聚类方法
CN102300140A (zh) * 2011-08-10 2011-12-28 歌尔声学股份有限公司 一种通信耳机的语音增强方法、装置及降噪通信耳机

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2680608A4

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106792315A (zh) * 2017-01-05 2017-05-31 歌尔科技有限公司 一种抵消环境噪声的方法和装置及一种主动降噪耳机
CN106792315B (zh) * 2017-01-05 2023-11-21 歌尔科技有限公司 一种抵消环境噪声的方法和装置及一种主动降噪耳机
CN107750028A (zh) * 2017-12-06 2018-03-02 贵州翔通科技实业有限公司 耳机
CN107750028B (zh) * 2017-12-06 2024-03-29 贵州翔通科技实业有限公司 耳机
CN111696564A (zh) * 2020-06-05 2020-09-22 北京搜狗科技发展有限公司 语音处理方法、装置和介质
CN111696564B (zh) * 2020-06-05 2023-08-18 北京搜狗科技发展有限公司 语音处理方法、装置和介质
CN112822592A (zh) * 2020-12-31 2021-05-18 青岛理工大学 一种可定向聆听的有源降噪耳机及控制方法
CN114007157A (zh) * 2021-10-28 2022-02-01 中北大学 一种智能降噪通信耳机

Also Published As

Publication number Publication date
US9484042B2 (en) 2016-11-01
JP2014507683A (ja) 2014-03-27
DK2680608T3 (en) 2016-04-25
KR101353686B1 (ko) 2014-01-20
CN102300140B (zh) 2013-12-18
CN102300140A (zh) 2011-12-28
EP2680608B1 (en) 2016-02-03
JP5513690B2 (ja) 2014-06-04
KR20130101152A (ko) 2013-09-12
US20140172421A1 (en) 2014-06-19
EP2680608A4 (en) 2014-10-22
EP2680608A1 (en) 2014-01-01

Similar Documents

Publication Publication Date Title
WO2013020380A1 (zh) 一种通信耳机的语音增强方法、装置及降噪通信耳机
WO2012069020A1 (zh) 语音增强方法、装置及头戴式降噪通信耳机
TWI763727B (zh) 使用多個麥克風的自動噪音消除
WO2011079722A1 (zh) 非封闭式耳塞型耳机及其受话端语音增强装置及方法
US8699719B2 (en) Personal acoustic device position determination
US8238567B2 (en) Personal acoustic device position determination
CN102348151B (zh) 噪声消除系统和方法、智能控制方法和装置、通信设备
EP2415276B1 (en) Personal acoustic device position determination
US9361902B2 (en) Earhole-wearable sound collection device, signal processing device, and sound collection method
KR20110099693A (ko) 스테레오 및 모노 신호를 재생하는 이어피스 및 방법
US11330358B2 (en) Wearable audio device with inner microphone adaptive noise reduction
WO2018147573A1 (ko) 인이어 마이크와 아웃이어 마이크 수음특성을 이용한 소음 제거 이어셋 및 소음 제거 방법
WO2011132845A2 (ko) 이어마이크폰
WO2018230880A1 (ko) 골전도 스피커를 내장한 이어셋
CN114503602A (zh) 用于耳戴式播放设备的音频系统和信号处理方法
JP2017011754A (ja) 耳孔装着型収音装置、信号処理装置、収音方法
JPH0879873A (ja) 通信端末
CN114450745A (zh) 用于耳戴式播放设备的音频系统和信号处理方法
WO2023283285A1 (en) Wearable audio device with enhanced voice pick-up
WO2024039049A1 (ko) 전자 장치 및 그 제어 방법
WO2024085576A1 (ko) 마이크가 내장된 외부 기기와 연결되는 스테레오 타입의 디지털 히어링 디바이스 및 그 동작 방법
CN114007163A (zh) 一种tws耳机自适应人体医学听力保护系统及使用方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12822487

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2013552095

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20137021521

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2012822487

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 14110879

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE