CN113329288B - Bluetooth headset noise reduction method based on notch technology - Google Patents
Bluetooth headset noise reduction method based on notch technology Download PDFInfo
- Publication number
- CN113329288B CN113329288B CN202110477342.2A CN202110477342A CN113329288B CN 113329288 B CN113329288 B CN 113329288B CN 202110477342 A CN202110477342 A CN 202110477342A CN 113329288 B CN113329288 B CN 113329288B
- Authority
- CN
- China
- Prior art keywords
- signal
- power spectrum
- noise
- band
- interference
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000005516 engineering process Methods 0.000 title claims abstract description 9
- 238000001228 spectrum Methods 0.000 claims abstract description 96
- 230000005236 sound signal Effects 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000002452 interceptive effect Effects 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 7
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000012938 design process Methods 0.000 description 2
- 230000001629 suppression Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
Images
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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques 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
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques 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
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L2021/02161—Number of inputs available containing the signal or the noise to be suppressed
- G10L2021/02166—Microphone arrays; Beamforming
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R2460/00—Details of hearing devices, i.e. of ear- or headphones covered by H04R1/10 or H04R5/033 but not provided for in any of their subgroups, or of hearing aids covered by H04R25/00 but not provided for in any of its subgroups
- H04R2460/01—Hearing devices using active noise cancellation
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The invention relates to a noise reduction method of a Bluetooth headset based on a notch technology, wherein two non-directional microphones which are linearly arranged are arranged on the Bluetooth headset to form a microphone array, the space is divided into a target space sub-band for acquiring a target sound signal and a plurality of interference space sub-bands based on the microphone array, and a noise power spectrum of an input signal of a current frame is obtained according to the power spectrum of the input signal and the noise power spectrum of each interference space sub-band; and taking the noise power spectrum of the input signal and the power spectrum of the input signal as input, and performing voice enhancement on the input signal by adopting a preset voice enhancement algorithm. And a better noise reduction effect is obtained under the condition that hardware resources are not increased much.
Description
Technical Field
The invention relates to the technical field of voice signal processing, in particular to a noise reduction method of a Bluetooth headset based on a trapped wave technology.
Background
When people hold communication equipment for conversation, if the surrounding environment is too noisy, the conversation quality can be seriously influenced by larger background noise, especially when the Bluetooth headset is used, because the microphone is far away from a speaker, the influence of the background noise on the conversation quality is more serious, and therefore, the noise reduction scheme of the Bluetooth headset is always paid attention.
As for noise reduction schemes of bluetooth headsets, there are many kinds of noise reduction schemes, single microphone, dual microphone and triple microphone, and some of them are added with bone conduction microphone. The two-microphone noise reduction scheme is relatively common, the two-microphone generally adopts a beam forming method, one is a noise reduction scheme based on adaptive beam forming, and the other is a noise reduction scheme based on fixed beam forming. Although the noise reduction scheme based on the adaptive beamforming has a good noise reduction effect, hardware resources need to be added, so that the cost is high, the space is increased, a user can easily cause large voice loss due to incorrect posture when wearing the bluetooth headset, and although the noise reduction scheme based on the fixed beamforming has low requirements on the hardware resources and higher robustness to the deviation of the speaking direction of a wearer, the noise reduction scheme based on the fixed beamforming has limited interference noise suppression capability and has poorer noise reduction effect as the distance from the speaking direction of the wearer is closer.
Disclosure of Invention
The invention provides a noise reduction method of a Bluetooth headset based on a notch technology, and aims to solve the technical problem that in the noise reduction scheme of the Bluetooth headset in the prior art, either the requirement on hardware resources is high or the noise reduction effect is poor.
A noise reduction method of a Bluetooth headset based on a notch technology is characterized in that a first microphone and a second microphone which are linearly arranged are arranged on the Bluetooth headset, and the first microphone and the second microphone form a microphone array; the method for dividing the space into a target space sub-band and a plurality of interference space sub-bands of a target sound signal based on a microphone array comprises the following steps:
Step a1, receiving an original sound signal using a microphone array;
step A2, dividing the original sound signal received by the first microphone according to a preset time interval and converting the original sound signal into a first frequency domain signal, dividing the original sound signal received by the second microphone according to the preset time interval and converting the original sound signal into a second frequency domain signal, and processing the first frequency domain signal and the second frequency domain signal to obtain a plurality of frame input signals of the microphone array;
step A3, calculating the power spectrum of the input signal based on the input signal;
step A4, adopting a notch wave beam of each interference space sub-band designed in advance to filter an input signal, and obtaining a filtered signal of each interference space sub-band;
step A5, calculating the noise power spectrum of each interference space sub-band based on the power spectrum of the input signal and the filtered signal of the interference space sub-band;
step A6, calculating the noise power spectrum of the current frame input signal based on the input signal power spectrum and the noise power spectrum of the interference space sub-band;
and step A7, taking the power spectrum of the input signal and the noise power spectrum of the input signal as input, and performing voice enhancement processing on the input signal by adopting a preset voice enhancement algorithm.
Step A8, converting the input signal after the speech enhancement of the current frame into a time domain signal and outputting the time domain signal;
and step A9, in the noise reduction method of the Bluetooth headset, circularly executing the steps A3-A8 to respectively perform voice enhancement processing on each frame of input signals and convert the input signals into time domain signals, and then combining all the frame time domain signals to be output as target sound signals after noise reduction so as to realize the noise reduction processing of the Bluetooth headset.
Further, in step a4, the design process of the notch beam of the interference spatial subband includes:
step A41, constructing an expression of a microphone array receiving signal in a frequency domain;
step A42, obtaining the estimated signal through a linear filter, and adopting the following expression:
Z(ω)＝hH(ω)Y(ω)＝hH(ω)d(ω)X(ω)+hH(ω)v(ω)；
step a43, setting a constraint condition to ensure that the target principal direction angle is undistorted, the constraint condition is as follows:
hH(ω)d(ω，θ)＝1；
step a44, setting constraints of the stop band beams to satisfy the constraints, so as to obtain the notch beams of the interference spatial sub-bands, where the constraints are as follows:
minh(ω)hH(ω)[εIM+Γα,β(ω)]h(ω)＝1；
wherein,
z (ω) is the estimation signal;
x (ω) is the desired signal;
h (omega) is a trapped wave beam to be solved;
the superscript H is a conjugate transpose operation;
y (omega) is an expression of a microphone array receiving signal in a frequency domain;
d (omega) is a space steering vector when angular frequency omega is taken as a variable to participate in calculation;
v (ω) is the noise signal vector;
d (ω, θ) is a spatial steering vector including the angular frequency ω and the incident angle θ as variables for calculation;
Γα,β(ω) is a uniform sound field noise covariance matrix for sound incidence angles in the range α to β;
epsilon is the white noise gain controlling the formation of the trapped wave beam;
IMis an identity matrix of order M.
Further, in step a41, the expression of the microphone array received signal in the frequency domain is as follows:
Y(ω)＝[Y1(ω)Y2(ω)]T＝x(ω)+v(ω)＝d(ω)X(ω)+v(ω)；
wherein,
Y1(ω) is an expression designed in the frequency domain for the received signal of the first microphone:
Y2(ω) is an expression designed in the frequency domain for the received signal of the first microphone:
x (ω) is the desired signal;
v (ω) is a noise signal vector.
Further, in step a4, the formula of the notch beam of the interference spatial subband is as follows:
wherein h isα,β(ω, θ) is a processing result of the notch beam of the interference spatial subband;
d (omega, theta) is a space guide vector when angular frequency omega and incidence angle theta are taken as variables to participate in calculation;
epsilon is the white noise gain controlling the formation of the trapped wave beam;
IMan identity matrix of order M;
Γα,β(ω) is a uniform sound field noise covariance matrix within a range of sound incidence angles α to β;
The superscript H is a conjugate transpose operation;
d (ω) is a spatial steering vector when an angular frequency ω is included as a variable in calculation.
Alpha represents a lower critical value of an incident angle range in an interference space sub-band;
beta represents the upper critical value of the incident angle range in the interference spatial subband.
Further, in step a4, the equation of the covariance matrix of the uniform sound field noise in the range of the incident angle α to β is as follows:
wherein,
Γα,β(ω) is a uniform sound field noise covariance matrix within a range of sound incidence angles α to β;
the superscript H is a conjugate transpose operation;
alpha represents a lower critical value of an incidence angle range in the interference space sub-band;
beta represents the upper critical value of the range of angles of incidence in the interfering spatial sub-bands.
Further, in step a5, the step of calculating the noise power spectrum of each interference space subband based on the power spectrum of the input signal and the filtered signal of the interference space subband includes the following steps:
step A51, calculating the power spectrum of the filtering signal of each interference space sub-band based on the filtered signal of the interference space sub-band;
step A52, calculating the noise power spectrum of each interference space sub-band based on the input signal power spectrum and the filtering signal power spectrum.
Further, in step a6, the step of calculating the noise power spectrum of the current frame input signal includes:
Step A61, calculating the signal-to-noise ratio of each interference space sub-band based on the power spectrum of the input signal and the noise power spectrum of the interference space sub-band;
step A62, calculating the noise existence probability of each interference space sub-band according to the signal-to-noise ratio of each interference space sub-band;
step A63, calculating a steady-state noise power spectrum according to an input signal;
and step A64, estimating the noise power spectrum of the input signal of the current frame according to the steady-state noise power spectrum and the noise power spectrum of each interference space sub-band by combining the noise existence probability of each interference space sub-band.
Further, in step a61, the snr of the interference spatial subband is calculated as follows:
wherein,
Girepresenting the signal-to-noise ratio of the ith interfering spatial subband;
nfft is the frequency point number of the single frame input signal;
Ni(ωn) The noise power spectrum of the ith interference space sub-band when the number of the frequency points is n;
P(ωn) Representing the power spectrum of the input signal with the number of frequency points n.
Further, in step a62, the calculation formula of the noise existence probability of the interference spatial subband is as follows:
wherein,
Peithe noise existence probability of the ith interference space sub-band;
Githe signal-to-noise ratio for the ith interfering spatial subband.
Further, the space is divided into a target space sub-band and three interference space sub-bands;
The angle range of the target space sub-band is 0-45 degrees, and the angle of the target main direction is 0 degree;
the angular ranges of the interference spatial sub-bands are 45 DEG to 90 DEG, 90 DEG to 135 DEG and 135 DEG to 180 DEG, respectively.
The beneficial technical effects of the invention are as follows: the invention discloses a Bluetooth headset double-microphone noise reduction method based on a trapped wave technology, which is used for noise power spectrum estimation by combining a method of noise power spectrum estimation of each spatial sub-band, can realize directional noise source tracking, obtains better noise reduction effect and increases little hardware resources.
Drawings
Fig. 1 is a schematic diagram of the arrangement of two microphones on the bluetooth headset of the present invention;
FIG. 2 is a schematic diagram of the main direction angles of sound signals received by the two microphones of the present invention;
FIG. 3 is a schematic diagram of a notch beam with an interference spatial subband angle range of 90-135 in accordance with the present invention;
FIGS. 4-5 are schematic flow charts of a method for reducing noise of a Bluetooth headset based on a notch technique according to the present invention;
fig. 6-9 are flowcharts illustrating steps of a method for reducing noise of a bluetooth headset according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
Referring to fig. 1-9, the present invention provides a noise reduction method for a bluetooth headset based on a notch technology, wherein a first microphone and a second microphone are arranged on the bluetooth headset in a linear arrangement, and the first microphone and the second microphone form a microphone array; based on a microphone array, the space is divided into a target space sub-band and a plurality of interference space sub-bands of a target sound signal, and the method comprises the following steps:
step a1, receiving an original sound signal using a microphone array;
step A2, dividing the original sound signal received by the first microphone according to a preset time interval and converting the original sound signal into a first frequency domain signal, dividing the original sound signal received by the second microphone according to the preset time interval and converting the original sound signal into a second frequency domain signal, and processing the first frequency domain signal and the second frequency domain signal to obtain a plurality of frame input signals of the microphone array;
step A3, calculating the power spectrum of the input signal based on the input signal;
Step A4, adopting a notch wave beam of each interference space sub-band designed in advance to filter an input signal, and obtaining a filtered signal of each interference space sub-band;
step A5, calculating the noise power spectrum of each interference space sub-band based on the power spectrum of the input signal and the filtered signal of the interference space sub-band;
step A6, calculating the noise power spectrum of the current frame input signal based on the input signal power spectrum and the noise power spectrum of the interference space sub-band;
step A7, using the power spectrum of the input signal and the noise power spectrum of the input signal as input, and performing speech enhancement processing on the input signal by adopting a preset speech enhancement algorithm.
Step A8, converting the input signal after the speech enhancement of the current frame into a time domain signal and outputting the time domain signal;
and step A9, in the noise reduction method of the Bluetooth headset, circularly executing the steps A3-A8 to respectively perform voice enhancement processing on each frame of input signals and convert the input signals into time domain signals, and then combining all the frame time domain signals to be output as target sound signals after noise reduction so as to realize the noise reduction processing of the Bluetooth headset.
Specifically, the first microphone and the second microphone are both non-directional microphones.
Specifically, the method carries out environmental noise suppression through four steps of notch beam former design, space subband noise power spectrum estimation, noise estimation and voice enhancement.
Specifically, the speech enhancement algorithm may adopt MMSE, spectral subtraction, and other algorithms.
Further, in step a4, the design process of the notch beams of the interference spatial subband includes:
step A41, constructing an expression of a microphone array receiving signal in a frequency domain;
further, the expression of the microphone array receiving signal in the frequency domain is as follows:
Y(ω)＝[Y1(ω) Y2(ω)]T＝x(ω)+v(ω)＝d(ω)X(ω)+v(ω)；
wherein,
Y1(ω) is an expression designed in the frequency domain for the received signal of the first microphone:
Y2(ω) is an expression designed in the frequency domain for the received signal of the first microphone:
x (ω) is the desired signal;
v (ω) is a noise signal vector.
Step A42, obtaining the estimated signal through a linear filter, and adopting the following expression:
Z(ω)＝hH(ω)Y(ω)＝hH(ω)d(ω)X(ω)+hH(ω)v(ω)；
step a43, setting a constraint condition to ensure that the target principal direction angle is undistorted, the constraint condition is as follows:
hH(ω)d(ω，θ)＝1；
step a44, setting the constraint conditions of the stop band beams to satisfy the constraint conditions, so as to obtain the above-mentioned notch beams h of the interference spatial sub-bandsα,β(ω, θ), the constraints are as follows:
minh(ω)hH(ω)[εIM+Γα,β(ω)]h(ω)＝1；
wherein,
z (omega) is an estimation signal;
x (ω) is the desired signal;
h (omega) is a trapped wave beam to be solved;
the superscript H is a conjugate transpose operation;
y (omega) is an expression of a microphone array receiving signal in a frequency domain;
d (omega) is a space steering vector when angular frequency omega is taken as a variable to participate in calculation;
v (ω) is the noise signal vector;
d (ω, θ) is a spatial steering vector including the angular frequency ω and the incident angle θ as variables for calculation;
Γα,β(omega) is soundA uniform sound field noise covariance matrix with a sound incidence angle ranging from alpha to beta;
epsilon is the white noise gain controlling the formation of the trapped wave beam;
IMis an identity matrix of order M.
Further, in step a4, the formula of the notch beam of the interference spatial subband is as follows:
wherein h isα,β(ω, θ) is a processing result of the notch beam of the interference spatial subband;
d (omega, theta) is a space guide vector when angular frequency omega and incidence angle theta are taken as variables to participate in calculation;
epsilon is the white noise gain controlling the formation of the trapped wave beam;
IMan identity matrix of order M;
Γα,β(ω) is a uniform sound field noise covariance matrix within a range of sound incidence angles α to β;
the superscript H is a conjugate transpose operation;
d (ω) is a spatial steering vector when an angular frequency ω is included as a variable to participate in the calculation.
Alpha represents a lower critical value of an incidence angle range in the interference space sub-band;
beta represents the upper critical value of the range of angles of incidence in the interfering spatial sub-bands.
Specifically, the expression of the spatial steering vector d (ω, θ) is as follows:
wherein,
d (ω, θ) is used to represent a steering vector when the incident angle is θ;
superscript T denotes a transpose operation;
ω ═ 2 π f for angular frequency;
f is the frequency;
t0＝d/c；
theta is an incident angle;
d is the spacing between the first microphone and the second microphone;
c is the speed of sound;
c＝340m/s。
further, in step a4, the equation of the covariance matrix of the noise of the uniform sound field with the incidence angle ranging from α to β is as follows:
wherein,
Γα,β(ω) is a uniform sound field noise covariance matrix within a range of sound incidence angles α to β;
the superscript H is a conjugate transpose operation;
alpha represents a lower critical value of an incidence angle range in the interference space sub-band;
beta represents the upper critical value of the range of angles of incidence in the interfering spatial sub-bands.
Specifically, the notch beam can ensure the target main direction angle, has no response distortion, and suppresses the directional sound field in the range from alpha to beta.
Specifically, in step a3, the power spectrum of the input signal is obtained by using the following formula:
P(ω)＝(pow(Y1(ω))+pow(Y2(ω)))/2，
wherein,
p (ω) is used to represent the input signal power spectrum;
pow is a power spectrum calculation formula;
Y1(ω) is an expression of the received original speech signal of the first microphone in the frequency domain;
Y2(ω) is an expression of the received original speech signal of the second microphone in the frequency domain.
The power spectrum of the input signal can be obtained through the formula.
Specifically, pow is a power spectrum calculation formula, pow (Y)1(ω))＝Y1(ω)*conj(Y1(ω))，pow(Y2(ω))＝Y2(ω)*conj(Y2(ω))。
Further, in step a5, the step of calculating the noise power spectrum of each interference space subband based on the power spectrum of the input signal and the filtered signal of the interference space subband includes the following steps:
step A51, calculating the power spectrum of the filtering signal of each interference space sub-band based on the filtered signal of the interference space sub-band;
step A52, calculating the noise power spectrum of each interference space sub-band based on the input signal power spectrum and the filtering signal power spectrum.
Specifically, in step a4, the expression of the filtered signal for each interfering spatial subband is as follows:
wherein,
since the already designed notch beam input signal is used to perform filtering processing in each interference space subband, Z is now the casei(ω) represents the filtered signal of the ith interfering spatial subband;
superscript H represents the conjugate transpose operation;
hia notch beam representing an ith interference space subband design;
y (omega) is an input signal.
Specifically, in step a51, the expression of the filtered signal power spectrum of the interference spatial subband is as follows:
Ei(ω)＝pow(Zi(ω))＝Zi(ω)*conj(Zi(ω))；
Wherein,
i＝1，2，3，…，m；
m represents the number of interference spatial subbands;
Ei(ω) is the filtered signal power spectrum of the ith interference space subband;
pow is a power spectrum calculation formula;
Zi(ω) represents the filtered signal of the ith interfering spatial subband.
Specifically, in step a52, the calculation formula of the noise power spectrum of each interference spatial subband is as follows:
Ni(ω)＝P(ω)-Ei(ω)；
wherein,
Ni(ω) is the noise power spectrum of the ith interfering spatial subband;
Ei(ω) is the filtered signal power spectrum of the ith interference spatial subband;
p (ω) is used to represent the input signal power spectrum.
Specifically, the power spectrum of the filtered signal of each interference spatial subband is expressed in a determinant manner as follows:
E(ω)＝[E1(ω)E2(ω)…Em(ω)]T；
the noise power spectrum of each interfering spatial subband is expressed in a determinant manner as follows:
N(ω)＝[N1(ω)N2(ω)…Nm(ω)]T＝P(ω)-E(ω)；
the superscript T denotes the transpose operation.
Further, in step a6, the step of calculating the noise power spectrum of the current frame input signal includes:
step A61, calculating the signal-to-noise ratio of each interference space sub-band based on the power spectrum of the input signal and the noise power spectrum of the interference space sub-band;
step A62, calculating the noise existence probability of each interference space sub-band according to the signal-to-noise ratio of each interference space sub-band;
step A63, calculating a steady-state noise power spectrum according to an input signal;
And step A64, estimating the noise power spectrum of the input signal of the current frame according to the steady-state noise power spectrum and the noise power spectrum of each interference space sub-band by combining the noise existence probability of each interference space sub-band.
Further, in step a61, the snr of the interference spatial subband is calculated as follows:
wherein,
i＝1，2，3，…，m；
Girepresenting the signal-to-noise ratio of the ith interfering spatial subband;
nfft is the frequency point number of the single frame input signal;
Ni(ωn) The noise power spectrum of the ith interference space sub-band when the number of the frequency points is n;
P(ωn) Representing the power spectrum of the input signal with the number of frequency points n.
Specifically, the snr of each interference spatial subband is expressed as a determinant:
G＝[G1G2…Gm]T；
further, in step a62, the calculation formula of the noise existence probability of the interference spatial subband is as follows:
wherein,
i＝1，2，3，…，m；
Peithe noise existence probability of the ith interference space sub-band;
Githe signal-to-noise ratio for the ith interfering spatial subband.
The noise existence probability of each interference space sub-band is expressed in a determinant mode as follows:
Pe＝[Pe1Pe2…Pem]T；
wherein,
m represents the number of interference spatial subbands;
the superscript T denotes the transpose operation.
Specifically, in step a64, the calculation formula of the noise power spectrum of the input signal is as follows:
Wherein,
NS (ω) represents the steady state noise power spectrum;
n (ω) is a determinant expression of the noise power spectrum of the interfering spatial sub-bands;
Ni(ω) is the noise power spectrum of the ith interfering spatial subband;
Peithe noise existence probability of the ith interference space sub-band;
NN (ω) is the noise power spectrum of the input signal.
Further, the space is divided into a target space sub-band and three interference space sub-bands;
the angle range of the target space sub-band is 0-45 degrees, and the angle of the target main direction is 0 degree;
the angle ranges of each interference spatial sub-band are 45-90 degrees, 90-135 degrees and 135-180 degrees respectively.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (10)
1. A noise reduction method of a Bluetooth headset based on a notch technology is characterized in that a first microphone and a second microphone which are linearly arranged are arranged on the Bluetooth headset, and the first microphone and the second microphone form a microphone array; based on the microphone array, the space is divided into a target space sub-band and a plurality of interference space sub-bands of a target sound signal, and the method comprises the following steps:
Step a1, receiving a raw sound signal using the microphone array;
step A2, dividing the original sound signal received by the first microphone according to a preset time interval and converting the original sound signal into a first frequency domain signal, dividing the original sound signal received by the second microphone according to a preset time interval and converting the original sound signal into a second frequency domain signal, and processing the first frequency domain signal and the second frequency domain signal to obtain a plurality of frame input signals of the microphone array;
step A3, calculating an input signal power spectrum based on the input signal;
step A4, performing filtering processing on the input signal by using a pre-designed notch beam of each interference space subband to obtain a filtered signal of each interference space subband;
step A5, calculating the noise power spectrum of each interference space sub-band based on the power spectrum of the input signal and the filtered signal of the interference space sub-band;
step A6, calculating the noise power spectrum of the input signal of the current frame based on the power spectrum of the input signal and the noise power spectrum of the interference space sub-band;
step A7, taking the power spectrum of the input signal and the noise power spectrum of the input signal as input, and performing voice enhancement processing on the input signal by adopting a preset voice enhancement algorithm;
Step A8, converting the input signal after the speech enhancement of the current frame into a time domain signal and outputting the time domain signal;
and step A9, in the noise reduction method of the Bluetooth headset, circularly executing the steps A3-A8 to respectively perform voice enhancement processing on each frame of input signals and convert the input signals into time domain signals, and then combining all the frame time domain signals to be output as the target sound signals after noise reduction so as to realize the noise reduction processing of the Bluetooth headset.
2. The method as claimed in claim 1, wherein in step a4, the designing process of the notch beam of the interference spatial subband comprises:
step A41, constructing an expression of a microphone array receiving signal in a frequency domain;
step A42, obtaining the estimated signal through a linear filter, and adopting the following expression:
Z(ω)＝hH(ω)Y(ω)＝hH(ω)d(ω)X(ω)+hH(ω)v(ω)；
step a43, setting a constraint condition to ensure that the target principal direction angle is undistorted, the constraint condition is as follows:
hH(ω)d(ω，θ)＝1；
step a44, setting constraints of stop band beams to satisfy the constraints, so as to obtain the notch beams of the interference spatial sub-bands, where the constraints are as follows:
minh(ω)hH(ω)[εIM+Γα，β(ω)]h(ω)＝1；
wherein,
z (omega) is an estimation signal;
x (ω) is the desired signal;
h (ω) is the notch beam to be solved;
the superscript H is a conjugate transpose operation;
y (omega) is an expression of a microphone array receiving signal in a frequency domain;
d (omega) is a space guide vector when angular frequency omega is used as a variable to participate in calculation;
v (ω) is the noise signal vector;
d (omega, theta) is a space guide vector when angular frequency omega and incidence angle theta are taken as variables to participate in calculation;
Γα，β(ω) is a uniform sound field noise covariance matrix with sound incidence angles in the range of α to β;
epsilon is the white noise gain controlling the formation of the trapped wave beam;
IMis an identity matrix of order M。
3. The method as claimed in claim 2, wherein in step a41, the expression of the microphone array received signal in the frequency domain is as follows:
Y(ω)＝[Y1(ω) Y2(ω)]T＝x(ω)+v(ω)＝d(ω)X(ω)+v(ω)；
wherein,
Y1(ω) is an expression designed in the frequency domain for the received signal of the first microphone:
Y2(ω) is an expression designed in the frequency domain for the received signal of the first microphone:
x (ω) is the desired signal;
v (ω) is a noise signal vector.
4. The method of claim 1, wherein in the step a4, the formula of the notch beam of the interference space sub-band is as follows:
Wherein is hα，β(ω, θ) is a processing result of the notch beam of the interference spatial subband;
d (ω, θ) is a spatial steering vector including the angular frequency ω and the incident angle θ as variables for calculation;
epsilon is the white noise gain for controlling the formation of the trapped wave beam;
IMan identity matrix of order M;
Γα，β(ω) is a uniform sound field noise covariance matrix within a range of sound incidence angles α to β;
the superscript H is a conjugate transpose operation;
d (omega) is a space guide vector when angular frequency omega is used as a variable to participate in calculation;
a represents a lower critical value of the incidence angle range in the interference space sub-band;
β represents an upper critical value of the range of angles of incidence in the interfering spatial sub-bands.
5. The method of claim 4, wherein in the step A4, the equation of the covariance matrix of the noise in the uniform sound field with the incidence angle from α to β is as follows:
wherein,
Γα，β(ω) is a uniform sound field noise covariance matrix within a range of sound incidence angles α to β;
the superscript H is a conjugate transpose operation;
a represents a lower critical value of the incidence angle range in the interference space sub-band;
β represents an upper critical value of the range of angles of incidence in the interfering spatial sub-bands.
6. The method of claim 1, wherein the step a5 of calculating the noise power spectrum of each of the interference spatial sub-bands based on the power spectrum of the input signal and the filtered signal of the interference spatial sub-band comprises the steps of:
step A51, calculating the power spectrum of the filtering signal of each interference space sub-band based on the filtered signal of the interference space sub-band;
step a52, calculating the noise power spectrum of each interference space sub-band based on the input signal power spectrum and the filtered signal power spectrum.
7. The method as claimed in claim 1, wherein the step of calculating the noise power spectrum of the input signal of the current frame in step a6 comprises:
step A61, calculating the signal-to-noise ratio of each interference space sub-band based on the power spectrum of the input signal and the noise power spectrum of the interference space sub-band;
step A62, calculating the noise existence probability of each interference space sub-band according to the signal-to-noise ratio of each interference space sub-band;
step A63, calculating a steady-state noise power spectrum according to the input signal;
Step a64, estimating the noise power spectrum of the input signal of the current frame according to the steady-state noise power spectrum and the noise power spectrum of each interference spatial sub-band by combining the noise existence probability of each interference spatial sub-band.
8. The method of claim 7, wherein in the step A61, the SNR of the interference spatial subband is calculated as follows:
wherein,
Girepresenting the signal-to-noise ratio of the ith interfering spatial subband;
nfft is the frequency point number of the input signal of a single frame;
Ni(ωn) The noise power spectrum of the ith interference space sub-band when the number of the frequency points is n;
P(ωn) Representing the power spectrum of the input signal with the number of frequency points n.
9. The method of claim 8, wherein in the step a62, the noise existence probability of the interference spatial sub-band is calculated as follows:
wherein,
Peithe noise existence probability of the ith interference space sub-band;
Githe signal-to-noise ratio for the ith interfering spatial subband.
10. The method of claim 1, wherein the space is divided into one target spatial sub-band and three interfering spatial sub-bands;
The angle range of the target space sub-band is 0-45 degrees, and the angle of the target main direction is 0 degree;
the angle ranges of each interference spatial sub-band are 45 DEG-90 DEG, 90 DEG-135 DEG and 135 DEG-180 deg respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110477342.2A CN113329288B (en) | 2021-04-29 | 2021-04-29 | Bluetooth headset noise reduction method based on notch technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110477342.2A CN113329288B (en) | 2021-04-29 | 2021-04-29 | Bluetooth headset noise reduction method based on notch technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113329288A CN113329288A (en) | 2021-08-31 |
| CN113329288B true CN113329288B (en) | 2022-07-19 |
Family
ID=77414105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110477342.2A Expired - Fee Related CN113329288B (en) | 2021-04-29 | 2021-04-29 | Bluetooth headset noise reduction method based on notch technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113329288B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114333888B (en) * | 2021-12-30 | 2025-02-11 | 北京声加科技有限公司 | Multi-beam joint noise reduction method and device based on white noise gain control |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010091077A1 (en) * | 2009-02-03 | 2010-08-12 | University Of Ottawa | Method and system for a multi-microphone noise reduction |
| CN102456351A (en) * | 2010-10-14 | 2012-05-16 | 清华大学 | Voice enhancement system |
| CN102306496B (en) * | 2011-09-05 | 2014-07-09 | 歌尔声学股份有限公司 | Noise elimination method, device and system of multi-microphone array |
| CN102509552B (en) * | 2011-10-21 | 2013-09-11 | 浙江大学 | Method for enhancing microphone array voice based on combined inhibition |
| CN105244036A (en) * | 2014-06-27 | 2016-01-13 | 中兴通讯股份有限公司 | Microphone speech enhancement method and microphone speech enhancement device |
| CN106504763A (en) * | 2015-12-22 | 2017-03-15 | 电子科技大学 | Multi-target Speech Enhancement Method Based on Microphone Array Based on Blind Source Separation and Spectral Subtraction |
| CN108694957B (en) * | 2018-04-08 | 2021-08-31 | 湖北工业大学 | Echo Cancellation Design Method Based on Circular Microphone Array Beamforming |
| CN110534125A (en) * | 2019-09-11 | 2019-12-03 | 清华大学无锡应用技术研究院 | A kind of real-time voice enhancing system and method inhibiting competitive noise |
| CN111968659B (en) * | 2020-07-23 | 2023-10-31 | 天津大学 | Microphone array voice enhancement method based on optimized IMCRA |
-
2021
- 2021-04-29 CN CN202110477342.2A patent/CN113329288B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN113329288A (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10535362B2 (en) | Speech enhancement for an electronic device | |
| US10224016B2 (en) | Voice sensing using multiple microphones | |
| US7983907B2 (en) | Headset for separation of speech signals in a noisy environment | |
| KR101519768B1 (en) | Method, device and system for eliminating noises with multi-microphone array | |
| CN110856072B (en) | Earphone conversation noise reduction method and earphone | |
| US20220189497A1 (en) | Bone conduction headphone speech enhancement systems and methods | |
| CN113544775B (en) | Audio signal enhancement for head-mounted audio devices | |
| US9589572B2 (en) | Stepsize determination of adaptive filter for cancelling voice portion by combining open-loop and closed-loop approaches | |
| CN108597532A (en) | Hidden method for acoustic based on MVDR | |
| CN113329288B (en) | Bluetooth headset noise reduction method based on notch technology | |
| Xu et al. | Adaptive speech enhancement algorithm based on first-order differential microphone array | |
| US11153695B2 (en) | Hearing devices and related methods | |
| Priyanka et al. | Adaptive Beamforming Using Zelinski-TSNR Multichannel Postfilter for Speech Enhancement | |
| EP3614696B1 (en) | Beam former, beam forming method and hearing aid system | |
| CN111327984B (en) | Earphone auxiliary listening method based on null filtering and ear-worn equipment | |
| CN117219108A (en) | Self-adaptive noise reduction method based on second-order differential microphone array | |
| Sunohara et al. | Low-latency real-time blind source separation with binaural directional hearing aids | |
| As’ad et al. | Robust minimum variance distortionless response beamformer based on target activity detection in binaural hearing aid applications | |
| Lotter et al. | A stereo input-output superdirective beamformer for dual channel noise reduction. | |
| US20250088793A1 (en) | Wearable audio devices with enhanced voice pickup | |
| CN116320947B (en) | A Frequency-Domain Dual-Channel Speech Enhancement Method Applied to Hearing Aids | |
| EP3886463B1 (en) | Method at a hearing device | |
| US20250088794A1 (en) | Wearable audio devices with enhanced voice pickup | |
| The et al. | A Spatial Spectral Mask for Dual-Microphone Systems | |
| The et al. | An Improved MVDR Beamformer's Performance by Using Dual-Microphone System |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220719 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |