CN110211601B - Method, device and system for acquiring parameter matrix of spatial filter - Google Patents
Method, device and system for acquiring parameter matrix of spatial filter Download PDFInfo
- Publication number
- CN110211601B CN110211601B CN201910425432.XA CN201910425432A CN110211601B CN 110211601 B CN110211601 B CN 110211601B CN 201910425432 A CN201910425432 A CN 201910425432A CN 110211601 B CN110211601 B CN 110211601B
- Authority
- CN
- China
- Prior art keywords
- matrix
- microphone
- spatial filter
- sub
- band frequency
- 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
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- 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
Abstract
The embodiment of the invention relates to the technical field of data processing, and particularly discloses a method, a device and a system for acquiring a parameter matrix of a spatial filter, wherein the method comprises the following steps: determining a first directional matrix under the first sub-band frequency according to a first sub-band frequency matrix under the pre-acquired first sub-band frequency, the number of microphones, the position of each microphone and the angle corresponding to the position of each microphone; determining element values in a constraint matrix according to the angle corresponding to the position of each microphone, wherein the number of the elements in the constraint matrix is the same as the number of the microphones; and determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix. By the aid of the method, the microphone array can be ensured to receive sound in the preset direction while short distance setting of the microphone array is met, and voice enhancement is achieved.
Description
Technical Field
The embodiment of the invention relates to the technical field of data processing, in particular to a method, a device and a system for acquiring a parameter matrix of a spatial filter and a storage medium.
Background
Beam forming requires controlling the direction of a beam by the time delay between microphones, and thus, an ideal shape is formed by beams with different directions. Thereby determining the direction from which beamforming wants to receive sound. In this way, it is achieved that some sound in a particular direction is received in its entirety or partially, and some sound in some direction is suppressed in its entirety. Generally speaking, the total suppressed sound is "noise" relative to the received sound and therefore needs to be suppressed.
If it is desired that the time delay between the microphones be strictly controlled according to a preset time delay, a certain distance between the microphones is required. However, once the microphone arrays are applied to a device with a smaller volume of earphones, the spacing between the microphone arrays cannot be arranged in a conventional manner due to the volume limitation of the device itself, so as to realize beam forming. The direct time delay of a short-spaced microphone array will be inaccurate or almost zero. That is, if the microphone array is arranged according to the conventional beamforming principle, the design of many hardware devices must be severely limited.
Then, in order to ensure that the short-distance setting of the microphone array is realized in a smaller volume range and that the microphone array can realize the beam forming in a preset state, receive the sound in a preset direction, suppress the noise in other directions, and achieve the speech enhancement, the most important step is to determine an optimal spatial filter parameter matrix, and the setting of the spatial filter parameter matrix directly affects the setting of the microphone array and further affects the beam forming, so how to obtain the optimal spatial filter parameter matrix becomes the technical problem to be solved by the application.
Disclosure of Invention
Therefore, embodiments of the present invention provide a method, an apparatus, a system, and a storage medium for obtaining a spatial filter parameter matrix, so as to solve the problem that in the prior art, an optimal beamforming spatial filter parameter matrix cannot be obtained, and thus, short-distance arrangement of a microphone array cannot be achieved in a small volume range, and meanwhile, beamforming in a preset state of the microphone array can be achieved, sound in a preset direction is received, noise in other directions is suppressed, and speech enhancement is achieved.
In order to achieve the above object, the embodiments of the present invention provide the following technical solutions:
in a first aspect, an embodiment of the present invention provides a method for obtaining a spatial filter parameter matrix, where the method includes:
determining a first directional matrix under the first sub-band frequency according to a first sub-band frequency matrix under the pre-acquired first sub-band frequency, the number of microphones, the position of each microphone and the angle corresponding to the position of each microphone;
determining element values in a constraint matrix according to the angle corresponding to the position of each microphone, wherein the number of the elements in the constraint matrix is the same as the number of the microphones;
and determining a spatial filter parameter matrix at a first subband frequency according to the constraint matrix and the first orientation matrix, wherein the number of the subbands is at least one, and the first subband is any one subband in the at least one subband.
The embodiment of the present invention is further characterized in that, according to a first sub-band frequency matrix under a pre-acquired first sub-band frequency, the number of microphones, a position of each microphone, and an angle corresponding to the position of each microphone, determining a first directional matrix under the first sub-band frequency specifically includes:
determining the time delay of the sound field reaching the position of each microphone under the first sub-band frequency according to the position of each microphone;
determining an ith pointing vector according to the first sub-band frequency matrix, the time delay when the sound field reaches the position of each microphone, and the angle corresponding to the position of the ith microphone, wherein i is initially taken as 1, and i is sequentially taken as values in a progressive manner until i is taken as a value corresponding to the number of the microphones;
from all the directional vectors, a first directional matrix at a first sub-band frequency is determined.
The embodiment of the invention is also characterized in that the angle corresponding to the position of each microphone comprises a pointing direction type and a restraining direction type; determining element values in a constraint matrix according to an angle corresponding to the position of each microphone, specifically comprising:
when the angle corresponding to the position of the first microphone is determined to be the type of the pointing direction, setting the element value corresponding to the first microphone in the constraint matrix to be 1; or, when the angle corresponding to the position of the first microphone is determined to be the suppression direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 0.
The embodiment of the present invention is further characterized in that the determining the spatial filter parameter matrix at the first subband frequency according to the constraint matrix and the first directional matrix specifically includes:
setting an initialization matrix of parameters of a spatial filter;
generating a cost function according to the first orientation matrix, the constraint matrix and the initialization matrix of the parameters of the spatial filter;
and solving the optimal solution of the cost function, and taking the parameter matrix of the spatial filter when the solution of the cost function is optimal as the final parameter matrix of the spatial filter.
The embodiment of the present invention is further characterized in that the optimal solution is solved for the cost function, and the spatial filter parameter matrix at the time of the optimal solution is used as a final spatial filter parameter matrix, which specifically includes:
and according to a Lagrange operator, carrying out convex optimization solution on the cost function to obtain an optimal solution, and taking the parameter matrix of the spatial filter when the optimal solution is obtained as a final parameter matrix of the spatial filter.
In a second aspect, an embodiment of the present invention further provides an apparatus for obtaining a spatial filter parameter matrix, where the apparatus includes:
the directional matrix determining unit is used for determining a first directional matrix under the first sub-band frequency according to a first sub-band frequency matrix under the first pre-acquired sub-band frequency, the number of the microphones, the position of each microphone and the angle corresponding to the position of each microphone;
the constraint matrix determining unit is used for determining element values in a constraint matrix according to the angle corresponding to the position of each microphone, wherein the number of the elements in the constraint matrix is the same as the number of the microphones;
and the processing unit is used for determining a spatial filter parameter matrix under a first subband frequency according to the constraint matrix and the first orientation matrix, wherein the number of the subbands is at least one, and the first subband is any one subband in the at least one subband.
The embodiment of the present invention is further characterized in that the directional matrix determining unit is specifically configured to determine, according to the position of each microphone, a time delay when the sound field reaches the position of each microphone at the first sub-band frequency;
determining an ith pointing vector according to the first sub-band frequency matrix, the time delay when the sound field reaches the position of each microphone, and the angle corresponding to the position of the ith microphone, wherein i is initially taken as 1, and i is sequentially taken as values in a progressive manner until i is taken as a value corresponding to the number of the microphones;
from all the directional vectors, a first directional matrix at a first sub-band frequency is determined.
The embodiment of the present invention is further characterized in that the angle corresponding to the position of each microphone includes a pointing direction type and a suppression direction type, and the constraint matrix determining unit is specifically configured to set the element value corresponding to the first microphone in the constraint matrix to 1 when it is determined that the angle corresponding to the position of the first microphone is the pointing direction type; or, when the angle corresponding to the position of the first microphone is determined to be the suppression direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 0.
In a third aspect, an embodiment of the present invention further provides a system for obtaining a spatial filter parameter matrix, where the system includes: a processor and a memory;
the memory is used for storing one or more program instructions;
a processor for executing one or more program instructions to perform any of the above method steps of the method for obtaining a spatial filter parameter matrix.
In a fourth aspect, an embodiment of the present invention further provides a computer storage medium, where the computer storage medium contains one or more program instructions, where the one or more program instructions are used by a server in a spatial filter parameter matrix acquisition system to perform any one of the method steps in the spatial filter parameter matrix acquisition method according to the first aspect.
According to the embodiment of the invention, the following advantages are provided: and determining a first directional matrix under the first sub-band frequency according to the pre-acquired sub-band frequency matrix under the first sub-band frequency, the number of the microphones, the position of each microphone and the angle corresponding to the position of each microphone. Then, determining element values in a constraint matrix according to the angle corresponding to the position of each microphone; and finally, determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix. The parameter matrix of the spatial filter is just an important parameter of beam forming, and the important parameter can meet the short-distance setting of the microphone array, and meanwhile, the microphone array can be ensured to realize the beam forming in a preset state, receive the sound in a preset direction, inhibit the noise in other directions, and achieve voice enhancement.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It should be apparent that the drawings in the following description are merely exemplary, and that other embodiments can be derived from the drawings provided by those of ordinary skill in the art without inventive effort.
Fig. 1 is a schematic flow chart of a method for obtaining a spatial filter parameter matrix according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an apparatus for obtaining a spatial filter parameter matrix according to another embodiment of the present invention;
fig. 3 is a schematic structural diagram of a system for obtaining a spatial filter parameter matrix according to another embodiment of the present invention.
Detailed Description
The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. 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.
An embodiment 1 of the present invention provides a method for obtaining a spatial filter parameter matrix, and specifically, as shown in fig. 1, the method may be applied to an earphone, VR glasses, or similar electronic devices with requirements on a multi-microphone array of a wearable device with strict requirements on volume. The method comprises the following steps:
Specifically, the acquired first sub-band frequency matrix at the first sub-band frequency, the number of microphones, the position of each microphone, and the angle corresponding to the position of each microphone are actually set according to the user requirement. In a specific example, the sound source frequency is 0-8000Hz, and the number of sub-bands can be set to 256. The first subband is any one of 256 subbands. In practical implementation, the time delay is different considering that the sound propagation speed is different in different frequency bands. Therefore, the spatial filter parameter matrix needs to be calculated for each subband. The manner of acquiring the first subband frequency matrix is prior art and will not be described herein.
The number of microphones, the positions of the microphones, and the corresponding angles are completely set according to actual conditions, for example, the number of microphones may be set to 6, and the microphone array forms a circle, and the positions of the microphones are 0 degree, 60 degree, 120 degree, 180 degree, 240 degree, 300 degree, and the like, so the angles corresponding to the positions of the microphones are natural, that is, the angles mentioned above.
Optionally, in the specific process of performing step 110, first, a time delay when the sound field reaches the position of each microphone under the first sub-band frequency needs to be determined according to the position of each microphone, and this process is a conventional technique and will not be described herein.
And then, determining the ith pointing vector according to the first sub-band frequency matrix, the time delay when the sound field reaches the position of each microphone, and the angle corresponding to the position of the ith microphone.
Specifically, see the following equation:
wherein d (w, θ)i) Is the ith vector of orientation, w isiThe angle corresponding to the position of the ith microphone,
is of TiFor the delay in the arrival of the sound source at the ith microphone, if the first microphone is taken as the reference microphone, then T1Is 0. M is the number of microphones, and in one specific example, M is 6. The dimensions of the pointing vector are 1 x 6 dimensions. Wherein, i is initially 1, i is orderlyAnd progressively taking values until the value of i is the value corresponding to the number of the microphones.
The first directional matrix is actually composed of a plurality of directional vectors, and at the first sub-band frequency, the number of the directional vectors is the same as the number of the microphones, and when the number of the microphones is 6, the number of the directional vectors is 6, so that the dimension of the first directional matrix is 6 × 6.
And step 120, determining element values in the constraint matrix according to the angle corresponding to the position of each microphone.
Specifically, the angles corresponding to the positions of the microphones include two types, wherein the first type is a pointing direction type; the second is the suppressed direction type. Then, determining the element values in the constraint matrix according to the angle corresponding to the position of each microphone, specifically including:
when the angle corresponding to the position of the first microphone is determined to be the type of the pointing direction, setting the element value corresponding to the first microphone in the constraint matrix to be 1;
or, when the angle corresponding to the position of the first microphone is determined to be the suppression direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 0.
In fact, each element in the constraint matrix represents a constraint, and more particularly represents a constraint on the angle at which the microphone is located. For example, when the angle corresponding to the position of the first microphone is the angle that the user wants to collect the sound, that is, the angle type is the pointing direction type, the elements corresponding to the positions in the constraint matrix are set to 1, and the angles corresponding to the positions of the second microphone to the sixth microphone are the suppression direction type, and the elements corresponding to the positions in the constraint matrix are set to 0. Naturally, the number of elements in the constraint matrix is the same as the number of microphones.
The constraint matrix expression at this time is as follows:
b ═ 100000' ("formula 2)
And step 130, determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix.
Specifically, a certain functional relationship exists between the first orientation matrix and the constraint matrix, and the functional relationship is related to the spatial filter parameter matrix. See equation 3 specifically:
loss function ═ a × h (w) -B (formula 3)
Wherein, A is the first directional matrix, B is the constraint matrix, H (w) is the spatial filter parameter matrix, and H (w) is the matrix that we require to get. The Loss function is a cost function, also called a Loss function.
Ideally, a × h (w) ═ B can be obtained only if the loss function is 0. However, in practical applications, the loss function cannot be zero, and then it needs to be minimized in an iterative process, and finally h (w) is obtained. In a specific implementation process, in order to implement the iterative process, an initialization matrix h (w) of spatial filter parameters may be set first. The elements in h (w) are randomly generated, and the dimension of h (w) is 6 x 1. And then, generating a cost function according to the first orientation matrix, the constraint matrix and the initialization matrix of the spatial filter parameters.
And solving the optimal solution of the cost function, and taking the parameter matrix of the spatial filter when the solution of the cost function is optimal as the final parameter matrix of the spatial filter.
Optionally, the convex optimization solution may be performed on the cost function according to a lagrangian operator to obtain an optimal solution, and the spatial filter parameter matrix when the optimal solution is obtained is used as the final spatial filter parameter matrix. The implementation process belongs to the prior art and is not described herein too much.
In addition, in the convex optimization solving process of the cost function, a limiting condition, namely white noise gain, can be additionally added, so that each element in the parameter matrix of the spatial filter of the filter can not amplify white noise, and further, the noise of a low frequency band can be prevented from being amplified.
After the above process is performed, the finally obtained filter spatial filter parameter matrix may be used to filter the audio signal collected by the microphone frame by frame, and the processed signal is input to the next noise reduction module for performing the subsequent operation.
It should be noted that the process of obtaining the optimal spatial filter parameter matrix by the above calculation may be performed by offline calculation, that is, a plurality of sets of filter spatial filter parameter matrices may be obtained in advance, so as to be adapted to the different requirements of the user. This design is not dependent on the electronics in which the microphone array is mounted, but can be implemented in any electronics with processing capabilities. The electronic equipment provided with the microphone array only needs to store the corresponding filter spatial filter parameter matrix, and does not need to store other data, namely, the requirements on the storage capacity and the calculation capacity of the electronic equipment provided with the microphone array can be properly reduced. Moreover, the method has obvious noise reduction effect on noisy noise and interference environment, and finally realizes the enhancement of the target voice.
According to the method for acquiring the parameter matrix of the spatial filter, provided by the embodiment of the invention, the first pointing matrix under the first sub-band frequency is determined according to the pre-acquired sub-band frequency matrix under the first sub-band frequency, the number of the microphones, the position of each microphone and the angle corresponding to the position of each microphone. Then, determining element values in a constraint matrix according to the angle corresponding to the position of each microphone; and finally, determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix. The parameter matrix of the spatial filter is just an important parameter of beam forming, and the important parameter can meet the short-distance setting of the microphone array, and meanwhile, the microphone array can be ensured to realize the beam forming in a preset state, receive the sound in a preset direction, inhibit the noise in other directions, and achieve voice enhancement.
Corresponding to the foregoing embodiment 1, an embodiment of the present invention further provides an apparatus for obtaining a spatial filter parameter matrix, specifically as shown in fig. 2, where the apparatus includes: a pointing matrix determination unit 201, a constraint matrix determination unit 202, and a processing unit 203.
A directional matrix determining unit 201, configured to determine a first directional matrix at a first sub-band frequency according to a first sub-band frequency matrix at the pre-acquired first sub-band frequency, the number of microphones, a position of each microphone, and an angle corresponding to the position of each microphone;
a constraint matrix determining unit 202, configured to determine element values in a constraint matrix according to an angle corresponding to a position of each microphone, where the number of elements in the constraint matrix is the same as the number of microphones;
and the processing unit 203 is configured to determine a spatial filter parameter matrix at a first subband frequency according to the constraint matrix and the first directional matrix, where the number of the subbands is at least one, and the first subband is any one of the at least one subband.
Optionally, the directional matrix determining unit 201 is specifically configured to determine, according to the position of each microphone, a time delay when the sound field reaches the position of each microphone at the first sub-band frequency;
determining an ith pointing vector according to the first sub-band frequency matrix, the time delay when the sound field reaches the position of each microphone, and the angle corresponding to the position of the ith microphone, wherein i is initially taken as 1, and i is sequentially taken as values in a progressive manner until i is taken as a value corresponding to the number of the microphones;
from all the directional vectors, a first directional matrix at a first sub-band frequency is determined.
Optionally, the angle corresponding to the position of each microphone includes a pointing direction type and a suppression direction type, and the constraint matrix determining unit 202 is specifically configured to, when it is determined that the angle corresponding to the position of the first microphone is the pointing direction type, set a value of an element corresponding to the first microphone in the constraint matrix to be 1; or, when the angle corresponding to the position of the first microphone is determined to be the suppression direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 0.
The functions performed by each component in the apparatus for obtaining a spatial domain filter parameter matrix according to the embodiment of the present invention are described in detail in the above embodiment 1, and therefore, redundant description is not repeated here.
According to the device for acquiring the parameter matrix of the spatial filter, provided by the embodiment of the invention, the first pointing matrix under the first sub-band frequency is determined according to the pre-acquired sub-band frequency matrix under the first sub-band frequency, the number of the microphones, the position of each microphone and the angle corresponding to the position of each microphone. Then, determining element values in a constraint matrix according to the angle corresponding to the position of each microphone; and finally, determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix. The parameter matrix of the spatial filter is just an important parameter of beam forming, and the important parameter can meet the short-distance setting of the microphone array, and meanwhile, the microphone array can be ensured to realize the beam forming in a preset state, receive the sound in a preset direction, inhibit the noise in other directions, and achieve voice enhancement.
Corresponding to the above embodiment 1, the embodiment of the present invention further provides a system for acquiring a spatial filter parameter matrix, specifically as shown in fig. 3, the system includes a processor 301 and a memory 302.
The memory 302 is used to store one or more program instructions;
the processor 301 is configured to execute one or more program instructions to perform any one of the above methods for obtaining a spatial filter parameter matrix.
The functions performed by each component in the system for acquiring a spatial domain filter parameter matrix according to the embodiment of the present invention are described in detail in embodiment 1 above, and therefore, redundant description is not repeated here.
According to the system for acquiring the parameter matrix of the spatial filter, provided by the embodiment of the invention, the first pointing matrix under the first sub-band frequency is determined according to the pre-acquired sub-band frequency matrix under the first sub-band frequency, the number of the microphones, the position of each microphone and the angle corresponding to the position of each microphone. Then, determining element values in a constraint matrix according to the angle corresponding to the position of each microphone; and finally, determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix. The parameter matrix of the spatial filter is just an important parameter of beam forming, and the important parameter can meet the short-distance setting of the microphone array, and meanwhile, the microphone array can be ensured to realize the beam forming in a preset state, receive the sound in a preset direction, inhibit the noise in other directions, and achieve voice enhancement.
In correspondence with the above embodiments, embodiments of the present invention also provide a computer storage medium containing one or more program instructions therein. Wherein, one or more program instructions are used for executing the spatial filter parameter matrix acquisition method described in embodiment 1 by a server in a spatial filter parameter matrix acquisition system.
Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (8)
1. A method for obtaining a spatial filter parameter matrix is characterized by comprising the following steps:
determining a first directional matrix under a first sub-band frequency according to a first sub-band frequency matrix under the first pre-acquired sub-band frequency, the number of microphones, the position of each microphone and the angle corresponding to the position of each microphone;
determining element values in a constraint matrix according to the angle corresponding to the position of each microphone, wherein the number of the elements in the constraint matrix is the same as the number of the microphones;
the angle corresponding to the position of each microphone comprises a pointing direction type and a suppression direction type; determining the element values in the constraint matrix according to the angle corresponding to the position of each microphone specifically includes:
when the angle corresponding to the position of the first microphone is determined to be a pointing direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 1;
or when the angle corresponding to the position of the first microphone is determined to be the suppression direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 0;
and determining a spatial filter parameter matrix under the first subband frequency according to the constraint matrix and the first orientation matrix, wherein the number of the subbands is at least one, and the first subband is any one subband in the at least one subband.
2. The method according to claim 1, wherein the determining, according to a first sub-band frequency matrix at a pre-acquired first sub-band frequency, the number of microphones, a position of each microphone, and an angle corresponding to the position of each microphone, a first directional matrix at the first sub-band frequency specifically includes:
determining the time delay of the sound field reaching the position of each microphone under the first sub-band frequency according to the position of each microphone;
determining an ith pointing vector according to the first sub-band frequency matrix, the time delay when the sound field reaches the position of each microphone, and the angle corresponding to the position of the ith microphone, wherein i is initially taken as 1, and i is sequentially taken as values in a progressive manner until i is taken as a value corresponding to the number of the microphones;
and determining a first pointing matrix under the first sub-band frequency according to all the pointing vectors.
3. The method according to any of claims 1-2, wherein determining the spatial filter parameter matrix at the first subband frequency based on the constraint matrix and the first orientation matrix comprises:
setting an initialization matrix of the parameters of the spatial filter;
generating a cost function according to the first orientation matrix, the constraint matrix and the initialization matrix of the parameters of the spatial filter;
and solving an optimal solution of the cost function, and taking the spatial filter parameter matrix when the solution of the cost function is optimal as a final spatial filter parameter matrix.
4. The method according to claim 3, wherein the solving of the optimal solution for the cost function and the taking of the spatial filter parameter matrix at the time of the optimal solution as a final spatial filter parameter matrix specifically comprises:
and according to a Lagrange operator, carrying out convex optimization solution on the cost function to obtain an optimal solution, and taking the parameter matrix of the spatial filter when the optimal solution is obtained as a final parameter matrix of the spatial filter.
5. An apparatus for obtaining a spatial filter parameter matrix, the apparatus comprising:
the directional matrix determining unit is used for determining a first directional matrix under the first sub-band frequency according to a first sub-band frequency matrix under the first pre-acquired sub-band frequency, the number of microphones, the position of each microphone and the angle corresponding to the position of each microphone;
a constraint matrix determining unit, configured to determine, according to an angle corresponding to a position of each microphone, an element number in a constraint matrix, where the element number in the constraint matrix is the same as the number of the microphones;
the angle corresponding to the position of each microphone includes a pointing direction type and a suppression direction type, and the constraint matrix determination unit is specifically configured to set a numerical value of an element corresponding to the first microphone in the constraint matrix to be 1 when it is determined that the angle corresponding to the position of the first microphone is the pointing direction type; or when the angle corresponding to the position of the first microphone is determined to be the suppression direction type, setting the element value corresponding to the first microphone in the constraint matrix to be 0;
a processing unit, configured to determine a spatial filter parameter matrix at the first subband frequency according to the constraint matrix and the first directional matrix, where the number of subbands is at least one, and the first subband is any one of the at least one subband.
6. The apparatus according to claim 5, wherein the directional matrix determining unit is specifically configured to determine, according to the location of each microphone, a delay time when the sound field reaches the location of each microphone at the first sub-band frequency;
determining an ith pointing vector according to the first sub-band frequency matrix, the time delay when the sound field reaches the position of each microphone, and the angle corresponding to the position of the ith microphone, wherein i is initially taken as 1, and i is sequentially taken as values in a progressive manner until i is taken as a value corresponding to the number of the microphones;
and determining a first pointing matrix under the first sub-band frequency according to all the pointing vectors.
7. A system for obtaining a spatial filter parameter matrix, the system comprising: a processor and a memory;
the memory is to store one or more program instructions;
the processor, configured to execute the one or more program instructions, to perform the method of any of claims 1-4.
8. A computer storage medium comprising one or more program instructions for execution by a processor in a spatial filter parameter matrix acquisition system to perform the method steps of any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910425432.XA CN110211601B (en) | 2019-05-21 | 2019-05-21 | Method, device and system for acquiring parameter matrix of spatial filter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910425432.XA CN110211601B (en) | 2019-05-21 | 2019-05-21 | Method, device and system for acquiring parameter matrix of spatial filter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110211601A CN110211601A (en) | 2019-09-06 |
| CN110211601B true CN110211601B (en) | 2020-05-08 |
Family
ID=67787951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910425432.XA Active CN110211601B (en) | 2019-05-21 | 2019-05-21 | Method, device and system for acquiring parameter matrix of spatial filter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110211601B (en) |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB0906269D0 (en) * | 2009-04-09 | 2009-05-20 | Ntnu Technology Transfer As | Optimal modal beamformer for sensor arrays |
| CN102214464B (en) * | 2010-04-02 | 2015-02-18 | 飞思卡尔半导体公司 | Transient state detecting method of audio signals and duration adjusting method based on same |
| US9100734B2 (en) * | 2010-10-22 | 2015-08-04 | Qualcomm Incorporated | Systems, methods, apparatus, and computer-readable media for far-field multi-source tracking and separation |
| WO2012086834A1 (en) * | 2010-12-21 | 2012-06-28 | 日本電信電話株式会社 | Speech enhancement method, device, program, and recording medium |
| EP2701145B1 (en) * | 2012-08-24 | 2016-10-12 | Retune DSP ApS | Noise estimation for use with noise reduction and echo cancellation in personal communication |
| CN106100608B (en) * | 2015-05-01 | 2019-02-05 | 中国人民解放军海军大连舰艇学院 | Weighted least-squares airspace matrix filter design method |
| CN108447499B (en) * | 2018-04-18 | 2020-08-04 | 佛山市顺德区中山大学研究院 | Double-layer circular-ring microphone array speech enhancement method |
| CN108717495A (en) * | 2018-05-22 | 2018-10-30 | 出门问问信息科技有限公司 | The method, apparatus and electronic equipment of multi-beam beam forming |
| CN109102822B (en) * | 2018-07-25 | 2020-07-28 | 出门问问信息科技有限公司 | Filtering method and device based on fixed beam forming |
| CN109616136B (en) * | 2018-12-21 | 2020-01-17 | 出门问问信息科技有限公司 | Adaptive beam forming method, device and system |
-
2019
- 2019-05-21 CN CN201910425432.XA patent/CN110211601B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110211601A (en) | 2019-09-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8981994B2 (en) | Processing signals | |
| US7809145B2 (en) | Ultra small microphone array | |
| US9253567B2 (en) | Array microphone apparatus for generating a beam forming signal and beam forming method thereof | |
| CN103000185B (en) | Processing signals | |
| US9020163B2 (en) | Near-field null and beamforming | |
| US11651772B2 (en) | Narrowband direction of arrival for full band beamformer | |
| WO2013085605A1 (en) | Near-field null and beamforming | |
| JP2005253071A (en) | System and method for beamforming using microphone array | |
| EP2749042A2 (en) | Processing signals | |
| CN110660404B (en) | Voice communication and interactive application system and method based on null filtering preprocessing | |
| KR102514060B1 (en) | A method of beamforming sound for driver units in a beamforming array and sound apparatus | |
| CN110211601B (en) | Method, device and system for acquiring parameter matrix of spatial filter | |
| EP3225037B1 (en) | Method and apparatus for generating a directional sound signal from first and second sound signals | |
| CN113223552B (en) | Speech enhancement method, device, apparatus, storage medium, and program | |
| JP6567216B2 (en) | Signal processing device | |
| CN110661510B (en) | Beam former forming method, beam forming device and electronic equipment | |
| CN113491137B (en) | Flexible differential microphone array with fractional order | |
| US20210241744A1 (en) | Directional noise suppression | |
| CN106448693A (en) | Speech signal processing method and apparatus | |
| CN108735228A (en) | Voice Beamforming Method and system | |
| CN109308907B (en) | single channel noise reduction | |
| JP2010152107A (en) | Device and program for extraction of target sound | |
| CN112785997B (en) | Noise estimation method and device, electronic equipment and readable storage medium | |
| WO2019223650A1 (en) | Beamforming method, multi-beam forming method and apparatus, and electronic device | |
| Durney et al. | A Multi-channel Speech Enhancement Method Based on Subband Affine Projection Algorithm in Combination with Proposed Circular Nested Microphone Array |
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 |