CN107889002B - Neck ring bluetooth headset, the noise reduction system of neck ring bluetooth headset and noise-reduction method - Google Patents
Neck ring bluetooth headset, the noise reduction system of neck ring bluetooth headset and noise-reduction method Download PDFInfo
- Publication number
- CN107889002B CN107889002B CN201711033558.XA CN201711033558A CN107889002B CN 107889002 B CN107889002 B CN 107889002B CN 201711033558 A CN201711033558 A CN 201711033558A CN 107889002 B CN107889002 B CN 107889002B
- Authority
- CN
- China
- Prior art keywords
- noise
- signal
- voice
- unit
- mike
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 230000009467 reduction Effects 0.000 title claims abstract description 50
- 238000000034 method Methods 0.000 title claims description 23
- 230000000694 effects Effects 0.000 claims abstract description 24
- 238000009434 installation Methods 0.000 claims abstract description 5
- 238000001914 filtration Methods 0.000 claims description 69
- 230000002596 correlated effect Effects 0.000 claims description 24
- 230000005236 sound signal Effects 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 10
- 230000001629 suppression Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 6
- 238000007493 shaping process Methods 0.000 claims description 6
- 241000209140 Triticum Species 0.000 claims description 4
- 235000021307 Triticum Nutrition 0.000 claims description 4
- 230000000875 corresponding effect Effects 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 6
- 238000001228 spectrum Methods 0.000 description 4
- 238000011946 reduction process Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- 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
- H04B—TRANSMISSION
- H04B5/00—Near-field transmission systems, e.g. inductive or capacitive transmission systems
-
- 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
- H04R1/08—Mouthpieces; Microphones; Attachments therefor
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Computer Networks & Wireless Communication (AREA)
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The present invention provides a kind of neck ring bluetooth headset, and the earphone includes the headset earpiece that the annular supporting structure with opening is connected with the earphone cable stretched out by the annular supporting structure；It is arranged in the middle part of the annular supporting structure towards outside auxiliary Mike, and the described second main Mike is installed towards the first inside main Mike or on the earphone cable in one end installation of annular supporting structure opening.Neck ring bluetooth headset of the present invention can effectively eliminate ambient noise and other people spoken sounds, reach better noise reduction effect, clean call environment more.
Description
Technical field
The present invention relates to bluetooth noise cancelling headphone technical field, the in particular to noise reduction of neck ring bluetooth headset, neck ring bluetooth headset
System and noise-reduction method.
Background technique
With the raising of social progress and living standards of the people, earphone has become essential life in for people's lives and uses
Product.People are usually in street, bus, and subway is answered and made a phone call, the noise in this kind of environment, can be by the call of earphone
The content that microphone is picked up and proximal end is spoken is mixed together, is sent in the earpiece of remote phone by uplink, is influenced distal end and is connect
The voice quality answered a call, therefore uplink noise reduction technology is particularly useful.Speech enhan-cement in call is big all the time
Amount concern the problem of, in noisy environment, how to allow distal end telephone user can not hear near-end speaker information to be expressed,
Inhibit noise, is an of great value research point.
Although typically now also can achieve preferable noise reduction effect, bluetooth using diamylose gram noise reduction schemes on mobile phone
The convenience of earphone makes many users be ready to be conversed or listened to music with earphone, so the noise reduction on earphone has very urgent need
It asks.Current common bluetooth headset does not have uplink decrease of noise functions substantially, and uplink call is generally using single Mike's decrease of noise functions, noise reduction
Ability is limited, only good to stationary noise effect, inhibits the effect of other noises in environment poor.In market, beaten in the environment such as subway
Phone is mingled with the sound of other tellers other than earphone user in addition to ambient noise in the sound that distal end telephone user hears
Sound, noise reduction effect are very poor.In the prior art, sound quality can be promoted due to being filtered to the useful signal in auxiliary microphone, however
Whole filtering can change the ambient noise in auxiliary microphone simultaneously, to cause the ambient noise signal and master in auxiliary microphone
Ambient noise signal difference in microphone is excessive, influences noise reduction effect.
Therefore, to solve the above-mentioned problems, it is desirable to be able to effectively eliminate personal spoken sounds around, reach better noise reduction
Effect makes the noise reduction system and noise-reduction method of the more clean neck ring bluetooth headset of call environment, neck ring bluetooth headset.
Summary of the invention
One aspect of the present invention is to provide a kind of neck ring bluetooth headset, and the earphone includes the annular branch with opening
The headset earpiece that support structure is connected with the earphone cable stretched out by the annular supporting structure；
It is arranged in the middle part of the annular supporting structure and is opened towards outside auxiliary Mike, and in the annular supporting structure
The second main Mike is installed in one end installation of mouth towards the first inside main Mike or on the earphone cable.
Preferably, one end of the annular supporting structure opening is also installed towards the inside main Mike of third.
Preferably, earphone cable is stretched out in the annular supporting structure two sides respectively.
Another aspect of the present invention is to provide a kind of noise reduction system of neck ring bluetooth headset, the system comprises:
Voice data input unit, for inputting the voice signal of main Mike's acquisition；Noise data input unit, for defeated
The noise signal for entering to assist Mike to acquire；
Whether speech activity detection unit, the voice signal for detecting the voice data input unit input come from people
Voice；Voice data judging unit, for judging whether the voice signal comes from earphone wearer；
Filter unit, the voice signal for sentencing unit judges using the voice data for wearer is to the noise number
Multi-subband time-domain filtering is carried out according to the noise signal of input unit input；
Noise suppression unit, for being adopted by the noise signal after the filter unit multi-subband time-domain filtering to main Mike
The voice signal of collection carries out noise suppressed.
Preferably, the system also includes noise reduction parameter computing units, for passing through the filter unit multi-subband
Noise signal after time-domain filtering updates noise reduction parameter.
Preferably, the system also includes beam shaping elements, the voice signal of first main Mike's acquisition passes through auxiliary wheat
Gram acquisition noise signal carry out noise suppressed generate first via output signal；
The voice signal of third main Mike acquisition carries out noise suppressed by the noise signal of auxiliary Mike's acquisition and generates the
Two tunnel output signals；
The beam shaping elements carry out beam forming processing to the first via output signal and the second tunnel output signal；
Or first main Mike acquisition voice signal and the voice signal of the main Mike's acquisition of third carry out Wave beam forming processing
Afterwards, noise suppressed is carried out by the noise signal of auxiliary Mike's acquisition.
Preferably, the voice data judging unit judges whether voice signal comes from earphone wearing by the following method
Person:
Delay process is carried out to noise signal, treated noise signal with voice signal is subjected to related for the first time locate
Reason, obtains the first correlated results；
Voice signal and noise signal are subjected to second of relevant treatment, obtain the second correlated results；
First correlated results and second correlated results are compared, if first correlated results is greater than institute
The second correlated results is stated, and the two difference is in preset threshold, then determines the voice signal from earphone wearer.
Preferably, the filter unit include the first time-domain filtering molecular band module, the second time-domain filtering molecular band module,
Multiple time domain filterings and adder, wherein
The first time-domain filtering molecular band module, for noise signal to be divided into multiple subbands；
The second time-domain filtering molecular band module, for the voice data to be sentenced the sound that unit judges are wearer
Signal is divided into multiple subbands；
The multiple time domain filtering sentences what the voice signal that unit judges are wearer divided for the voice data
Multiple subbands carry out time-domain filtering to multiple subbands that the noise signal divides respectively；
The synthesizer, for by the filtered subsignal of each time domain filtering in the multiple time domain filtering into
Row superposition.
Preferably, in multiple subbands that the voice signal that the voice data sentences that unit judges are wearer divides, frequency
Corresponding frequencies range in multiple subbands that each subband that range is in 300Hz~3.4KHz divides the noise signal
Subband carry out time-domain filtering.
Another aspect of the invention is to provide a kind of noise-reduction method of neck ring bluetooth headset, which comprises
Main Mike's collected sound signal, and voice signal is separately input by speech activity by voice data input unit
Detection unit and noise suppression unit；
Mike's acquisition noise signal is assisted, and noise signal is transmitted separately to by voice number by noise data input unit
It is judged that unit and filter unit；
Speech activity detection unit detect voice data input unit input voice signal whether the language from people
Sound, if
The voice signal is then input to voice data judging unit, the voice data judging unit passes through the sound
Sound signal and the noise signal judge whether the voice signal comes from earphone wearer, if
The transmission of sound signals to filter unit is then subjected to multi-subband time-domain filtering to the noise signal；
Noise suppression unit acquires main Mike by the noise signal after the filter unit multi-subband time-domain filtering
Voice signal carries out noise suppressed.
Neck ring formula bluetooth headset provided by the invention, by main Mike and auxiliary Mike by neck ring structure in remote point
It opens, keeps the voice signal of main Mike's acquisition more accurate, the judgement of wearer's spoken sounds can be effectively performed.
The noise reduction system of neck ring bluetooth headset provided by the invention, by the sound that phonetic decision unit judges wearer is arranged
Sound signal, distinguish other people around call in spoken sounds, what very good solution energy method diamylose gram noise reduction cannot achieve
The defect of single voice around wearer.
The noise reduction system of neck ring bluetooth headset provided by the invention, when carrying out multi-subband to noise signal by filter unit
Domain filtering, to balance sound quality and noise reduction effect.
It should be appreciated that aforementioned description substantially and subsequent detailed description are exemplary illustration and explanation, it should not
As the limitation to the claimed content of the present invention.
Detailed description of the invention
With reference to the attached drawing of accompanying, the more purposes of the present invention, function and advantage are by the as follows of embodiment through the invention
Description is illustrated, in which:
Fig. 1 diagrammatically illustrates the structural schematic diagram of neck ring bluetooth headset in one embodiment of the invention；
Fig. 2 shows the structural schematic diagrams of neck ring bluetooth headset in another embodiment of the present invention；
Fig. 3 shows the structural schematic diagram of neck ring bluetooth headset in further embodiment of the present invention；
Fig. 4 shows the overall structure block diagram of the noise reduction system of neck ring bluetooth headset of the present invention；
Fig. 5 shows the flow diagram that voice data judging unit of the present invention judges the voice signal of wearer；
Fig. 6 shows the flow diagram that filter unit of the present invention carries out multi-subband time-domain filtering to noise acoustical signal；
Fig. 7 shows the flow diagram of noise suppressed in voice signal of the present invention.
Specific embodiment
By reference to exemplary embodiment, the purpose of the present invention and function and the side for realizing these purposes and function
Method will be illustrated.However, the present invention is not limited to exemplary embodiment as disclosed below；Can by different form come
It is realized.The essence of specification is only to aid in those skilled in the relevant arts' Integrated Understanding detail of the invention.
Below by specific embodiment to the neck ring bluetooth headset of offer of the invention, the noise reduction system and noise reduction of neck ring bluetooth headset
Method provides detailed embodiment, and neck ring earphone, which refers to, to be conversed earphone by being placed on wearer's volume neck or music
Bluetooth headset.The structural schematic diagram of neck ring bluetooth headset in one embodiment of the invention as shown in Figure 1, neck ring in the present embodiment
Bluetooth headset includes connecting with the annular supporting structure 100 being open and the earphone cable 102 stretched out by annular supporting structure 100
Headset earpiece (not shown).Auxiliary Mike 103 towards outside (side B) is set in the middle part of annular supporting structure 100, and
First main Mike 101 of the installation towards inside (side A) in one end (end a) that annular supporting structure 100 is open.Wearer is by neck ring
Bluetooth headset is worn on neck, wearer's voice signal is acquired by the first main Mike 101, by assisting Mike 103
Acquire ambient noise.According to the present invention, main Mike and auxiliary Mike are passed through into neck ring structure in separating at a distance in embodiment, made
The voice signal of main Mike's acquisition is more accurate, the judgement of wearer's spoken sounds can be effectively performed, and to wearer
Voice signal carries out more efficiently noise reduction.
The structural schematic diagram of neck ring bluetooth headset in another embodiment of the present invention as shown in Figure 2, neck ring in the present embodiment
Bluetooth headset includes connecting with the annular supporting structure 100 being open and the earphone cable 102 stretched out by annular supporting structure 100
Headset earpiece (not shown).Auxiliary Mike 103 towards outside (side B) is set in the middle part of annular supporting structure 100, and
Second main Mike 104 is installed on earphone cable 102.Neck ring bluetooth headset is worn on neck by wearer, passes through the second main Mike
104 acquisition wearer's voice signals, by auxiliary, Mike 103 acquires ambient noise.It according to the present invention, will in embodiment
Main Mike with auxiliary Mike by neck ring structure at a distance separating, the voice signal for acquiring main Mike is more accurate, can be with
The judgement of wearer's spoken sounds is effectively performed, and more efficiently noise reduction is carried out to wearer's voice signal.
The structural schematic diagram of neck ring bluetooth headset in further embodiment of the present invention as shown in Figure 3, neck ring in the present embodiment
Bluetooth headset includes connecting with the annular supporting structure 100 being open and the earphone cable 102 stretched out by annular supporting structure 100
Headset earpiece (not shown).Auxiliary Mike 103 towards outside (side B) is set in the middle part of annular supporting structure 100, and
First main Mike 101 of the installation towards inside (side A) in one end (end a) that annular supporting structure 100 is open.In the present embodiment,
The main Mike 105 of third towards inside (side A) is also installed in one end (end a) that annular supporting structure 100 is open, and passes through the first main wheat
Grams 101 and the main Mike 105 of third acquire wearer's voice signal respectively, voice signal is divided into two-way and is carried out at noise reduction
Detailed description is given below in reason, specific noise reduction process.It carries out wearer and neck ring bluetooth headset is worn on neck, pass through
First main Mike 101 acquires wearer's voice signal, and by auxiliary, Mike 103 acquires ambient noise.According to the present invention,
Main Mike and auxiliary Mike are passed through into neck ring structure in separating at a distance in embodiment, make the voice signal of main Mike's acquisition more
Accurately, the judgement of wearer's spoken sounds can be effectively performed, and more efficiently drop is carried out to wearer's voice signal
It makes an uproar.
In some embodiments, earphone cable is stretched out in annular supporting structure two sides respectively, listens for the use of wearer's left-right ear
Cylinder.
The noise reduction of judgement to neck ring bluetooth headset wearer's spoken sounds in above-described embodiment and voice signal below
Process is described in detail, due to main Mike (the first main Mike, the second main Mike or the first main Mike and the main Mike of third)
When acquiring wearer's voice signal, ambient noise and surrounding other staff speech are inevitably collected
Sound, while Mike being assisted inevitably to collect wearer's voice signal.According to the present invention to above-mentioned noise
And wearer's surrounding people's voice signal inhibits, to reach the noise reduction to wearer's voice signal.
The present invention is illustrated for the noise reduction system of neck ring bluetooth headset first, neck ring bluetooth headset of the present invention as shown in Figure 4
The overall structure block diagram of noise reduction system, a kind of noise reduction system suitable for neck ring bluetooth headset in embodiment, comprising:
Voice data input unit 201, for inputting main Mike (the first main Mike, the second main Mike or the first main wheat
Gram and the main Mike of third) acquisition voice signal, noise data input unit 202, for input auxiliary Mike acquire noise
Signal.
Whether speech activity detection unit 203, the voice signal for detecting the input of voice data input unit 201 come from
The voice of people, voice data judging unit 204, for judging whether voice signal comes from earphone wearer.
Filter unit 205 is judged as the voice signal of wearer to noise data for sentencing unit 204 using voice data
The noise signal that input unit 202 inputs carries out multi-subband time-domain filtering.
The noise reduction system of neck ring bluetooth headset further includes noise reduction parameter computing unit 206, the sound of main Mike's acquisition
Signal and the filtered noise signal of filter unit 205 are first into noise reduction parameter computing unit 206, pass through filtering
Noise signal after unit multi-subband time-domain filtering updates noise reduction parameter.
Noise suppression unit 207, for being adopted by the noise signal after filter unit multi-subband time-domain filtering to main Mike
The voice signal of collection carries out noise suppressed.
The noise reduction system of neck ring bluetooth headset provided by the invention carries out noise reduction process by the following method, specifically a kind of
The noise-reduction method of neck ring bluetooth headset, which comprises
Main Mike's collected sound signal, and voice signal is separately input by voice by voice data input unit 201
Activity detector units 203 and noise suppression unit 207.
Mike's acquisition noise signal is assisted, and noise signal is transmitted separately to by language by noise data input unit 202
Sound data determining unit 204 and filter unit 205.
Whether speech activity detection unit 203 detects the voice signal of the input of voice data input unit 201 from people's
Voice, if
Voice signal is then input to voice data judging unit 204, voice data judging unit 204 passes through voice signal
Judge whether voice signal comes from earphone wearer with noise signal, if
Transmission of sound signals to filter unit 205 is then subjected to multi-subband time-domain filtering to noise signal；
If judging voice signal right and wrong from earphone wearer, this section of voice signal is as background noise.
The sound that noise suppression unit 207 acquires main Mike by the noise signal after filter unit multi-subband time-domain filtering
Sound signal carries out noise suppressed.
Judge whether voice signal comes from wearer
Voice data judging unit of the present invention as shown in Figure 5 judges the flow diagram of the voice signal of wearer, this hair
When bright speech activity detection unit 203 detects that the voice signal of main Mike's input is had from people's voice, voice number
It is judged that unit 204 judges whether voice signal comes from earphone wearer by the following method:
Delay process is carried out to noise signal, treated noise signal with voice signal is subjected to related for the first time locate
Reason, obtains the first correlated results, the first correlated results is calculated in the following way in embodiment:
Voice signal and noise signal are subjected to second of relevant treatment, the second correlated results is obtained, passes through in embodiment
As correlated results is calculated under type:
Wherein
X1 (k) is noise signal, and x2 (k) is voice signal, and coeff1 is the first correlated results, and coeff2 is second related
As a result, τ is the delay duration of the main Mike's voice signal of the first correlated results treatment process.
First correlated results and the second correlated results are compared, if the first correlated results is greater than the second correlated results,
And the two difference is in preset threshold, then determines the voice signal from earphone wearer.
The present invention is handled by voice data judging unit correlated results, is effectively sentenced to the sound of main Mike acquisition
It is disconnected, wearer is come from when the voice signal for determining main Mike's acquisition is non-, then as background noise by this section of voice signal.
Voice signal from wearer filters noise signal
Multi-subband time-domain filtering is carried out to noise signal by the voice signal from wearer, as shown in Figure 6 the present invention
Filter unit carries out the flow diagram of multi-subband time-domain filtering to noise signal, and according to the present invention, filter unit 205 includes the
One time-domain filtering molecular band module 301, the second time-domain filtering molecular band module 302, multiple time domain filterings (time domain filtering 1,
Time domain filtering 2 ..., time domain filtering N) and adder 303, wherein
First filtering molecular band module 301, for noise signal to be divided into multiple subbands；
Second filtering molecular band module 302 is judged as that the voice signal of wearer is drawn for voice data to be sentenced unit 204
It is divided into multiple subbands.
Multiple time domain filterings sentence multiple subbands that the voice signal that unit judges are wearer divides for voice data
Time-domain filtering is carried out to multiple subbands that the noise signal divides respectively.
Synthesizer, for the filtered subsignal of each time domain filtering in multiple time domain filterings to be overlapped.
Specifically, the noise signal x1 of input is divided into x11 by the first filtering molecular band module 301 in embodiment,
Voice data is sentenced the sound that unit 204 is judged as wearer by the more a subbands of x12 ..., x1N, the second filtering molecular band module 302
Signal x2 is divided into x21, the more a subbands of x22 ..., x2N.Multiple subband x21 of voice signal x2, x22 ..., x2N are respectively to making an uproar
Multiple subband x11 of acoustical signal x1, x12 ..., x1N are filtered in time domain filtering, it may be assumed that
X21 carries out time-domain filtering to x11 in time domain filtering 1；
X22 carries out time-domain filtering to x12 in time domain filtering 2；
X2N carries out time-domain filtering to x1N in time domain filtering N.
It should be noted that above-mentioned time-domain filtering process, voice data is sentenced the voice signal that unit judges are wearer and is drawn
Point multiple subbands in, multiple sons that each subband that frequency range is in 300Hz~3.4KHz divides noise signal
Subband in band within the scope of corresponding frequencies carries out time-domain filtering.
Each time domain filtering exports the subsignal of filtered noise signal, multiple subsignal z11, z12 ..., z1N respectively
It is overlapped by synthesizer device, the noise signal X1 after obtaining filtering unit filters.Preferably, time domain is multiple in the present embodiment
Time domain filtering uses normalization minimum mean-square error-filter (NLMS), calculates by the following method: z1n=NLMS (x1n,
X2n), wherein the number of n expression subband.
Filtered noise signal is synthesized by synthesizer, is specifically calculated as follows filtered
Noise signal: X1=z11+z12+ ...+z1N.
Since voice signal is mainly within the scope of 300Hz~3.4KHz, the present invention is filtered by molecular band, to auxiliary microphone
Voice signal in the noise signal of acquisition within the scope of certain frequency carries out adaptive voice filtering, removal auxiliary microphone acquisition
Noise in wearer voice signal, retain the voice signal of non-wearer to promote sound quality, for being higher than certain frequency
Signal does not filter to guarantee noise reduction effect.
Noise signal is to voice signal noise reduction
The present invention has been in by the processing of filtered noise signal above-described embodiment through the invention, noise signal
Clean noise signal (including ambient noise signal and non-wearer's voice signal), below by filtered noise
Signal carries out noise reduction process to voice signal, and the flow diagram of noise suppressed, implements in voice signal of the present invention as shown in Figure 7
Filtered noise signal X1 (k) and main Mike collected sound signal x2 (k) by adding window and carry out in quick Fu respectively in example
Leaf transformation (FFT), obtains frequency domain data:
Wherein m is frequency domain subscript, and n is time domain frame number, and L indicates frame step-length when there is overlapping between consecutive frame, and w is length
For the analysis window function (such as Hanning window) of K, X1(m, n), X2(m, n) is that X1 (k) is corresponded to after X2 (k) Short Time Fourier Transform
Frequency domain data, g be voice signal X2 gain correction factor.
Gain correction factor g is obtained by the ratio calculation of the standard deviation to the bottom noise cancellation signal received under two-way mute state
It arrives:
Wherein the standard deviation of input signal can pass through
It is calculated, wherein NN indicates to calculate the frame number of signal used in standard deviation, and u is the mean value of x (k).
According to the characteristic of mute section of calculating noise between speech interval, noise power spectrum is calculated by following iterative algorithm
Density Pw(m, n) is obtained
PW(m, n)=λW·PW(m,n-1)+(1-λW)·|X(m,n)|2If n < Nmax (4)
Wherein NmaxThe frame number maximum value of the close needs of noise power spectrum, λ are sought in expressionWIt is forgetting factor.It composes gain G (m, n)
For
WhereinThe power spectral density that Mike receives signal is calculate by the following formula
It obtains
PX(m, n)=λX·PX(m,n-1)+(1-λX)·|X(m,n)|2 (6)
H12(m, n) indicates the Fourier transformation of the shock response ratio between voice signal and noise signal, under meeting
Column equation
WhereinIndicate the power after noise is received by auxiliary Mike,WithIt respectively indicates auxiliary
The power for the noise signal for helping Mike and main Mike to receive,WithRespectively assist Mike and main Mike
The power of the voice signal received wherein assumes voice signal in embodiment and noise signal is independent.According to actual field
Scape, the voice signal power that voice signal power ratio another microphone received close to the microphone of sound source receives is more
By force, it is calculated by formula (4) | H12(m, n) |, it is applied to formula (5) and spectrum gain G (m, n) is calculated, G (m, n) indicates dry
The ratio of the power spectrum for the voice signal that net voice signal power spectral density and Mike receive.So available finally go
Except the frequency-region signal after noise
S (m, n)=G (m, n) X2(m,n) (8)
S (m, n) passes through Fourier inversion and superposition corresponding with frame step-length L, the voice signal exported:
Y (k)=ovlp (IFFT [S (m, n)])
Above-described embodiment inhibits the noise in voice signal by noise suppression unit, and the live streaming of output is made to put loudspeaker
Sound more cleans.
When neck ring blue-tooth earphone structure of the present invention is to have the first main 101 He of Mike in embodiment illustrated in fig. 3
It further include beam shaping elements in the noise reduction system of neck ring bluetooth headset when the main Mike 105 of third, the first main Mike 101 acquires
Voice signal by auxiliary Mike 103 acquire noise signal carry out noise suppressed generate first via output signal.
The voice signal that the main Mike 105 of third acquires carries out noise suppressed by the noise signal that auxiliary Mike 103 acquires
Generate the second tunnel output signal.
Beam shaping elements carry out beam forming processing to first via output signal and the second tunnel output signal, obtain final
The voice signal of output；
Or first main Mike acquisition voice signal and the voice signal of the main Mike's acquisition of third carry out Wave beam forming processing
Afterwards, noise suppressed is carried out by the noise signal of auxiliary Mike's acquisition
Neck ring formula bluetooth headset provided by the invention is laid out neck ring formula bluetooth headset microphone, by main Mike with
Mike is assisted to separate by neck ring structure in remote, two microphones are apart from larger, the data gain of two-way microphone acquisition
Difference is larger, for the diamylose gram noise reduction algorithm of use, can obtain preferable noise reduction effect.Make the sound of main Mike's acquisition simultaneously
Signal is more accurate, and the judgement of wearer's spoken sounds can be effectively performed.
The noise reduction system of neck ring bluetooth headset provided by the invention, by the sound that phonetic decision unit judges wearer is arranged
Sound signal, distinguishes other people around call in spoken sounds, judges whether voice is said by wearer using correlated results, with
Other single or multiple people talk the ambient noise to be formed around eliminating.It keeps silence in wearer, and it is close apart from wearer
In the range of have one or more other speaker, speech detection will be unable to judge the voice be wearer say or
The ambient noise of surrounding, this will affect diamylose gram noise reduction effect.Judge that the segment signal is useful voice letter using correlated results
Number or ambient noise, it will promoted diamylose gram noise reduction effect, effectively inhibit surrounding voice formed ambient noise.
The noise reduction system of neck ring bluetooth headset provided by the invention, when carrying out multi-subband to noise signal by filter unit
Domain filtering, to balance sound quality and noise reduction effect.By to the voice signal molecular band in certain band limits to auxiliary Mike
The collected noise signal of wind carries out adaptive temporal filter, and sound quality is promoted while obtaining preferable noise reduction effect.
In conjunction with the explanation and practice of the invention disclosed here, the other embodiment of the present invention is for those skilled in the art
It all will be readily apparent and understand.Illustrate and embodiment is regarded only as being exemplary, true scope of the invention and purport are equal
It is defined in the claims.
Claims (9)
1. a kind of noise reduction system of neck ring bluetooth headset, which is characterized in that the system comprises: earphone, including with opening
The headset earpiece that annular supporting structure is connected with the earphone cable stretched out by the annular supporting structure；
It is arranged in the middle part of the annular supporting structure towards outside auxiliary Mike, and in annular supporting structure opening
The second main Mike is installed in one end installation towards the first inside main Mike or on the earphone cable；Voice data input unit,
For inputting the voice signal of main Mike's acquisition；Noise data input unit, for inputting the noise signal of auxiliary Mike's acquisition；
Speech activity detection unit, for detect voice data input unit input voice signal whether the language from people
Sound；Voice data judging unit, for judging whether the voice signal comes from earphone wearer；
Filter unit, it is defeated to the noise data for sentencing the voice signal that unit judges are wearer using the voice data
The noise signal for entering unit input carries out multi-subband time-domain filtering；
Noise suppression unit, for what is acquired by the noise signal after the filter unit multi-subband time-domain filtering to main Mike
Voice signal carries out noise suppressed.
2. system according to claim 1, which is characterized in that also install direction in one end of the annular supporting structure opening
The main Mike of inside third.
3. system according to claim 1, which is characterized in that earphone cable is stretched out in the annular supporting structure two sides respectively.
4. system according to claim 1, which is characterized in that the system also includes noise reduction parameter computing unit,
For updating noise reduction parameter by the noise signal after the filter unit multi-subband time-domain filtering.
5. system according to claim 1, which is characterized in that the system also includes beam shaping elements, the first main wheat
Gram acquisition voice signal by auxiliary Mike acquisition noise signal carry out noise suppressed generate first via output signal；
The voice signal of the main Mike's acquisition of third carries out noise suppressed by the noise signal of auxiliary Mike's acquisition and generates the second tunnel
Output signal；
The beam shaping elements carry out beam forming processing to the first via output signal and the second tunnel output signal；
Or first main Mike acquisition voice signal and the main Mike's acquisition of third voice signal carry out Wave beam forming processing after, lead to
The noise signal for crossing auxiliary Mike's acquisition carries out noise suppressed.
6. system according to claim 1, which is characterized in that the voice data judging unit judges by the following method
Whether voice signal comes from earphone wearer:
Delay process is carried out to noise signal, treated noise signal and voice signal are subjected to first time relevant treatment, obtained
To the first correlated results；
Voice signal and noise signal are subjected to second of relevant treatment, obtain the second correlated results；
First correlated results and second correlated results are compared, if first correlated results is greater than described the
Two it is related as a result, and the two difference be in preset threshold, then determine that the voice signal comes from earphone wearer.
7. system according to claim 1, which is characterized in that the filter unit includes the first time-domain filtering molecular band mould
Block, the second time-domain filtering molecular band module, multiple time domain filterings and adder, wherein
The first time-domain filtering molecular band module, for noise signal to be divided into multiple subbands；
The second time-domain filtering molecular band module, for the voice data to be sentenced the voice signal that unit judges are wearer
It is divided into multiple subbands；
The multiple time domain filtering sentences the multiple of the voice signal division that unit judges are wearer for the voice data
Subband carries out time-domain filtering to multiple subbands that the noise signal divides respectively；
The adder, for folding the filtered subsignal of each time domain filtering in the multiple time domain filtering
Add.
8. system according to claim 1, which is characterized in that the voice data sentences the sound that unit judges are wearer
In multiple subbands that signal divides, frequency range is in each subband in 300Hz~3.4KHz and draws to the noise signal
The subband of corresponding frequencies range carries out time-domain filtering in the multiple subbands divided.
9. a kind of noise-reduction method of neck ring bluetooth headset, which is characterized in that the described method includes:
Main Mike's collected sound signal, and voice signal is separately input by voice activity detection by voice data input unit
Unit and noise suppression unit；
Mike's acquisition noise signal is assisted, and noise signal is transmitted separately to by voice data by noise data input unit and is sentenced
Disconnected unit and filter unit；
Speech activity detection unit detect voice data input unit input voice signal whether the voice from people, if
It is
The voice signal is then input to voice data judging unit, the voice data judging unit is believed by the sound
Number and the noise signal judge whether the voice signal comes from earphone wearer, if
The transmission of sound signals to filter unit is then subjected to multi-subband time-domain filtering to the noise signal；
The sound that noise suppression unit acquires main Mike by the noise signal after the filter unit multi-subband time-domain filtering
Signal carries out noise suppressed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711033558.XA CN107889002B (en) | 2017-10-30 | 2017-10-30 | Neck ring bluetooth headset, the noise reduction system of neck ring bluetooth headset and noise-reduction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711033558.XA CN107889002B (en) | 2017-10-30 | 2017-10-30 | Neck ring bluetooth headset, the noise reduction system of neck ring bluetooth headset and noise-reduction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107889002A CN107889002A (en) | 2018-04-06 |
| CN107889002B true CN107889002B (en) | 2019-08-27 |
Family
ID=61782881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711033558.XA Active CN107889002B (en) | 2017-10-30 | 2017-10-30 | Neck ring bluetooth headset, the noise reduction system of neck ring bluetooth headset and noise-reduction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107889002B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DK179837B1 (en) * | 2017-12-30 | 2019-07-29 | Gn Audio A/S | Microphone apparatus and headset |
| CN111131947B (en) * | 2019-12-05 | 2022-08-09 | 小鸟创新(北京)科技有限公司 | Earphone signal processing method and system and earphone |
| US11508391B2 (en) * | 2020-05-29 | 2022-11-22 | HearMe Technology LLC | Device to amplify and clarify voice |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1643571A (en) * | 2002-03-27 | 2005-07-20 | 艾黎弗公司 | Nicrophone and voice activity detection (vad) configurations for use with communication systems |
| CN102282865A (en) * | 2008-10-24 | 2011-12-14 | 爱利富卡姆公司 | Acoustic voice activity detection (avad) for electronic systems |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102859591B (en) * | 2010-04-12 | 2015-02-18 | 瑞典爱立信有限公司 | Method and arrangement for noise cancellation in a speech encoder |
| US9037458B2 (en) * | 2011-02-23 | 2015-05-19 | Qualcomm Incorporated | Systems, methods, apparatus, and computer-readable media for spatially selective audio augmentation |
| CN102300140B (en) * | 2011-08-10 | 2013-12-18 | 歌尔声学股份有限公司 | Speech enhancing method and device of communication earphone and noise reduction communication earphone |
| CN202773057U (en) * | 2012-08-10 | 2013-03-06 | 刘天镐 | Necklace type wireless earphone with noise reduction and magnetic therapy functions |
| RU2520184C1 (en) * | 2012-12-28 | 2014-06-20 | Алексей Леонидович УШАКОВ | Headset of mobile electronic device |
| CN207382519U (en) * | 2017-10-30 | 2018-05-18 | 恒玄科技（上海）有限公司 | A kind of neck ring bluetooth headset |
-
2017
- 2017-10-30 CN CN201711033558.XA patent/CN107889002B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1643571A (en) * | 2002-03-27 | 2005-07-20 | 艾黎弗公司 | Nicrophone and voice activity detection (vad) configurations for use with communication systems |
| CN102282865A (en) * | 2008-10-24 | 2011-12-14 | 爱利富卡姆公司 | Acoustic voice activity detection (avad) for electronic systems |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107889002A (en) | 2018-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108712703B (en) | The high-efficient noise-reducing earphone and noise reduction system of low-power consumption | |
| US8751224B2 (en) | Combined microphone and earphone audio headset having means for denoising a near speech signal, in particular for a “hands-free” telephony system | |
| CN107371079B (en) | A kind of the diamylose gram noise reduction system and noise-reduction method of earphone | |
| CN109195042B (en) | Low-power-consumption efficient noise reduction earphone and noise reduction system | |
| US8611554B2 (en) | Hearing assistance apparatus | |
| CN105229737B (en) | Noise cancelling microphone device | |
| JP6150988B2 (en) | Audio device including means for denoising audio signals by fractional delay filtering, especially for "hands free" telephone systems | |
| CN107533838A (en) | Sensed using the voice of multiple microphones | |
| CN107889002B (en) | Neck ring bluetooth headset, the noise reduction system of neck ring bluetooth headset and noise-reduction method | |
| KR20110107833A (en) | Acoustic in-ear detection for earpiece | |
| KR101647974B1 (en) | Smart earphone, appliance and system having smart mixing module, and method for mixing external sound and internal audio | |
| KR20090050372A (en) | Noise cancelling method and apparatus from the mixed sound | |
| CN104092801A (en) | Intelligent terminal call noise reduction method and intelligent terminal | |
| CN103813251B (en) | Hearing-aid denoising device and method allowable for adjusting denoising degree | |
| CN109151635A (en) | Realize the automatic switchover system and method that active noise reduction and the outer sound of ear are picked up | |
| CN109920451A (en) | Voice activity detection method, noise suppressing method and noise suppressing system | |
| US10972844B1 (en) | Earphone and set of earphones | |
| KR20100138804A (en) | Apparatus for enhancing intelligibility of speech, voice output apparatus with the apparatus | |
| CN105427868A (en) | Method for eliminating noise of VOIP system bidirectional duplex hand-free voice | |
| CN104994456A (en) | Earphone capable of improving conversation tone quality and method thereof | |
| US20220392475A1 (en) | Deep learning based noise reduction method using both bone-conduction sensor and microphone signals | |
| CN110931027A (en) | Audio processing method and device, electronic equipment and computer readable storage medium | |
| JP6548938B2 (en) | Speech processing apparatus and speech processing method | |
| CN116343756A (en) | Human voice transmission method, device, earphone, storage medium and program product | |
| CN207382519U (en) | A kind of neck ring bluetooth headset |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 201306 room 904, No. 800, Huanhu West 2nd Road, Lingang New Area, China (Shanghai) pilot Free Trade Zone, Pudong New Area, Shanghai Patentee after: Hengxuan Technology (Shanghai) Co., Ltd Address before: 200135 Shanghai 647 free trade pilot area, 171 Pudong New Area Road, 1 buildings, 6 rooms Patentee before: Heng Xuan Technology (Shanghai) Co., Ltd. |
|
| CP03 | Change of name, title or address |