Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
Implementation environment
Referring to fig. 1, fig. 1 is a schematic diagram of an implementation environment architecture according to an embodiment of the present invention.
As shown in fig. 1, includes: host 101, user 102, sound sources 103 to 109.
The user 102 is a person with hearing impairment or hearing deficiency, has the ability to swing the head freely, and can follow the instructions of the fitter after wearing the hearing aid, and when hearing the sound from the surrounding sound source, can look for the sound source by swinging the head (i.e., the face is facing the direction of the sound source).
The sound sources 103 to 109 are sound sources located around the user 102, and the cameras are mounted in the test environment, optionally on the sound sources or on top of the user. In a possible embodiment, the sound sources 103 to 109 are uniformly distributed on a circle with a radius of 3m and the user 102 is the center of the circle within 180 degrees in front of the user 102, and optionally, the sound sources 103, 106 and 109 are provided with cameras or the top of the user is provided with cameras.
It should be noted that, in order to ensure that the user completely judges the position of the sound source according to the hearing, the user 101 and the sound sources 103 to 109 need to use a barrier for visual isolation, which is a feasible measure as long as the user 101 cannot see the sound source without affecting the effect of hearing the sound emitted by the sound source, optionally, all the sound sources are placed on a support with a mesh enclosure or all the sound sources and the user 101 are directly isolated by using an opaque cloth, and the like, and in addition, in order to meet the requirement of a general life scene, the distance between the user 101 and the sound source in the test environment should be moderate, preferably, the distance is about 3 m.
The host 101 may be a desktop computer, a notebook, a palm computer, a cloud server, a smart phone, or other computing devices. The host computer 101 can control any sound source 103 to 109 to play sound, so that the user 102 can hear the sound. The host 101 is installed with corresponding debugging software, which can adjust the relevant parameters of the hearing aid, wherein the relevant parameters of the hearing aid mainly refer to gain parameters in the present invention.
The host computer 101 is connected to a hearing aid worn by the user 102 by wire or wirelessly. In one possible implementation, the host 101 and the user 102 are connected via bluetooth. In another possible embodiment, the host 101 and the user 102 are connected via a hearing aid programming line.
The host 101 and the sound emission sources 103 to 109 are connected by wire or wirelessly.
Those skilled in the art will appreciate that the foregoing embodiments are illustrative only, and are not intended to be limiting.
Fig. 2 is a schematic diagram of an embodiment of a method for adjusting the balance of a binaural hearing aid according to an embodiment of the present invention, including:
in step 201, a fitting condition for balance adjustment is provided, and the fitting condition includes: fixing a plurality of sound sources and determining the angles of the sound sources.
The proposed method of binaural hearing aid balance adjustment comprises certain implementation conditions, i.e. in a specific operation, the fitter needs to be equipped with a fixed sound source at the site where the hearing aid balance adjustment is to be made, where "fixed" emphasizes that the number and location of the sound sources should be known to be determined during the adjustment.
Optimally, the sounding sources are 7 in number, are spaced by 30 degrees and are uniformly distributed on the front semicircular arc of the user. The resolution of the fitting condition in each direction is accurate enough, the adjustment can be fully satisfied with the recognition degree of the sound direction in the daily actual scene, and the improvement brought by increasing the number of the sound sources for adjustment is very small.
All the sound sources constitute a test sound source group, and for the convenience of description, the test sound source group will be described below.
It should be noted that, the angles of the positions of any two sound sources in the test sound source group are different, for example, the sound sources 103 to 109 of the above-mentioned implementation environment are uniformly distributed and fixed on a circle with a radius of 3m and a 180-degree range around the user 102, which can sufficiently ensure the point, but in fact, the invention should not be limited to this.
In step 202, in a first step, a target sound source is controlled to play sound, wherein the target sound source is randomly selected.
The fitter needs to provide a sound source capable of emitting sound for the user to test during the balance adjustment of the binaural hearing aid. In an embodiment of the present invention, the sound sources may be selected as sound boxes, and the test sound source group is described as a test sound box group accordingly.
In order to ensure the reliability and accuracy of the balance adjustment of the binaural hearing aid, at least 2 sound sources are included in the test sound source group. In a possible implementation, 2 sound boxes in the test sound box group are all connected to the host, and can receive a play instruction from the host to play a specified sound file, and optionally, the format of the sound file includes WMA, MP3, MIDI, and the like. The host corresponds to the host 101 in fig. 1.
In practice, the testing engineer usually needs to test the adjustment repeatedly to complete the testing goal during the whole balancing process. The target sound source is a sound source which is controlled by the dispenser to play sound through the host computer in the current adjusting process, and the target sound source is randomly selected by the dispenser from devices which do not complete balance adjustment in the previous adjusting process.
In one possible implementation, the test sound source group has 3 sound boxes, which are numbered as number 1, number 2 and number 3 in sequence. In the first adjustment, 3 sound boxes are not tested, the testing and matching engineer can select one sound box of No. 1 or No. 2 or No. 3 to test, if the sound box of No. 1 is selected to finish the first adjustment, in the second adjustment, the testing and matching engineer can select one sound box of No. 2 or No. 3 to test, if the sound box of No. 3 is selected to finish the second adjustment, in the third adjustment, the testing and matching engineer selects the rest sound box of No. 2 to test.
In step 203, the second step, the head angle of the user is determined.
In step 202, after hearing the sound played by the target sound source, the user faces the face to the position of the target sound source under the instruction of the dispenser. The number of the cameras in this step 203 should be not less than 3, so that it can be ensured that the information of the head portrait of the user shot by the cameras can be sufficiently used for determining the head angle of the user.
In step 204, the third step, calculating an angle deviation according to the head angle of the user and the angle of the target sound source, and determining a target gain difference.
After determining the head angle of the user after hearing the sound, via step 203, a target gain difference may be determined that may help guide the hearing aid adjustment based on the deviation between the head angle of the user and the angle of the position of the device currently playing the sound.
In step 205, in the fourth step, according to the target gain difference, corresponding parameter adjustment is performed on the hearing aid worn by the user.
The target gain value determined in step 204 can help guide the adjustment of parameters of the hearing aid, so that the head angle of the user can be closer to the real angle of the target sound source after the adjustment of the hearing aid. The method of the first step, the second step, the third step and the fourth step is adopted for a plurality of times in sequence until the balance adjustment of the double-ear hearing aid is completed
The embodiment of the invention provides a method for adjusting the balance of a binaural hearing aid, which comprises the steps of firstly controlling a target sound source to play sound by a host, wherein the target sound source is randomly selected from a test sound source group by the host, then facing a face to the position of the target sound source under the instruction of an experienced fitter after the user hears the sound played by the target sound source, then determining a target gain difference according to the angle deviation between the head angle of the user and the target sound source, and finally performing corresponding parameter adjustment on the hearing aid worn by the user according to the target gain difference to complete the balance adjustment of the binaural hearing aid. Through the mode, the testing and matching technician can efficiently and conveniently guide the user with hearing impairment or hearing deficiency to complete the balance adjustment test of the binaural hearing aid, the aim of balance adjustment of the binaural hearing aid is fulfilled, and the user can more accurately distinguish the source of sound in daily work and life after wearing the adjusted hearing aid.
It is a feasible solution for the above-mentioned embodiments of the present invention to continuously perform the test by controlling the target sound source to play sound in a random manner in the first step, step 202, to finally complete the balance adjustment of the binaural hearing aid, but it necessarily requires a long test time period to ensure that all the sound sources are in balance.
Therefore, in order to further improve the efficiency of balance adjustment, in another embodiment of the present invention, in the above steps, the host controls the target loudspeaker to play sound, wherein the process of randomly selecting the target loudspeaker from the test loudspeaker group for the host is further optimized, and referring to fig. 3 and fig. 4, the following is described in detail:
fig. 3 is a schematic diagram of the operation of the host console screen during a binaural hearing aid balance adjustment. Fig. 4 is a flowchart of a method for controlling a target speaker to play sound by a host.
In the present embodiment, the sound source is selected as a sound box, in fig. 3, the host console screen 301 is an operation interface of the fitter during the balance adjustment of the hearing aid, where the host corresponds to the host 101 in fig. 1, and the sound boxes 302 to 308 together form a test sound box group.
The process depicted in FIG. 4 includes:
in step 401, grouping is performed according to the balance state of each sound source in the test sound source group, wherein the test sound source group is divided into a balance group and an unbalance group;
it should be noted that all sound sources in the balanced group are in a balanced state, and all sound sources in the unbalanced group are in an unbalanced state. The balance state is not the attribute of the sound source, but is used for measuring whether the sound source completes the test in the balance adjustment process of the binaural hearing aid, if the sound source does not complete the balance test, the sound source is in an unbalanced state, otherwise, the sound source is in a balanced state.
In step 402, a target sound source is randomly selected from an unbalanced set in a test sound source group;
it should be noted that the random selection described in the above step 402 is only for illustrating the difference of the selection sequence of the target sound source, and is not limited in the present invention, and is not intended to emphasize the uncertainty of the selection sequence.
As shown in fig. 3, the best implementation mode adopted in this embodiment is that the test sound box group has 7 sound boxes, and the numbers 1 to 7 shown on the screen 301 of the host console correspond to an entity sound box in the implementation environment, and when the balance state of the sound boxes changes in the fitting process, the balance state of the entity sound box can be distinguished through the background shape, that is, whether balance adjustment is completed or not. Specifically, the circular backgrounds of the sound boxes 302 and 303 represent that the physical sound boxes numbered 1 and 2 in the test sound box group have been balanced and are in a balanced state, while the other physical sound boxes numbered 3 to 7 in the test sound box group have not been balanced and are in an unbalanced state, and the background is square. In the test enclosure group, enclosure 1 and enclosure 2 form a balanced group and enclosures 3 through 7 form an unbalanced group. In the present binaural hearing aid balance adjustment process shown in fig. 3, the fitter controls the speakers 3 in the unbalanced group to play the specified sound file by clicking the speakers 3 on the screen of the host console, and at this time, the speakers 3 are the target speakers. It will be appreciated that the balanced and unbalanced groups in the test speaker cluster will change dynamically as the fitting process progresses.
Compared with the method of the previous embodiment, the method of the present embodiment further increases the process of controlling the target sound box to play sound by the fitter through the host, uniformly controls the test sound box group through the host console, and can conveniently manage the test progress through the visual interface, for example, the fitter can clearly know the number of sound boxes in a balanced state, that is, the number of sound boxes in a balanced group, according to the background shape of the corresponding entity sound box on the interface.
On the basis of the embodiment corresponding to fig. 2, the determining of the head angle of the user according to the embodiment of the present invention includes:
determining the head angle of the user based on the head portrait of the user shot by the camera, wherein the number of the cameras is one or more than one in different angles;
in the method disclosed by the invention, the number of the cameras is at least 1, the cameras are used for shooting the head portrait of the user in the test process, and the position of the camera needs to be capable of accurately shooting the head portrait of the user. In one possible implementation, the number of cameras is 1, which can be mounted on the top of the user's head or on any test sound source. In another possible implementation, the number of cameras is 3, and the cameras can be installed on the sound box 103, the sound box 106 and the sound box 109 in fig. 2, respectively. It should be noted that when the number of the cameras is larger, the head portrait of the user can be shot from more angles, and more head angle information of the user can be acquired, which is helpful for more accurately determining the head angle of the user.
Before determining the head angle of the user based on the head portrait of the user shot by the camera, the method may further include:
acquiring corresponding head photos shot by a camera at the test position of a user when the head is at different angles;
the method comprises the steps of extracting feature data of a head photo, and establishing a head angle judgment model by combining a head angle corresponding to the head photo through a data-driven modeling method, wherein an input parameter of the head angle judgment model is the feature data of the head photo, and an output parameter is the head angle.
It should be noted that in the method of performing balance adjustment during fitting, a very important step is to be able to determine the head angle of the user. The fitter is required to perform a large number of hearing aid fitting tasks for many different users each day, and the binaural hearing aid balance adjustment is only part of the fitting process. Therefore, the invention discloses a head angle judgment model acquisition method, and a dispenser can rapidly determine the head angle of a user in the process of fitting by using the model, so that the fitting efficiency is greatly improved.
The invention discloses a head angle judgment model obtaining method, which is based on a large number of sample head photos of a head at different angles at a test position of a user, obtains characteristic data of the head photos through a head portrait characteristic extraction algorithm, combines the head angles of the head photos to form a large number of sample data, and establishes a relation between the head portrait of the user and the user angle by adopting a data-driven modeling method. Optionally, the modeling method based on data driving may be a deep learning algorithm based on a Convolutional Neural Network (CNN), which will be taken as an illustration in the embodiments of the present invention, but the present invention is not limited to this method, and the modeling method based on data driving is feasible in the scheme proposed by the present invention.
This embodiment is described with reference to fig. 5 on the basis of the embodiment corresponding to fig. 3, and fig. 5 is a schematic diagram illustrating obtaining the head angle of the user.
Fig. 5 includes cameras 501 to 503, user avatars 504 to 506, the user avatar 504 captured by the camera 501, the user avatar 505 captured by the camera 502, and the user avatar 506 captured by the camera 503. The implementation environment has 7 sound boxes, and the cameras are installed in the sound boxes No. 1, No. 3 and No. 7, which correspond to the cameras 501 to 503 in fig. 5 in sequence.
First, a large number of different groups of head portrait sample data are collected, specifically, a group of sample data should include a user head portrait shot by a camera and a user head angle at the time, that is, an angle at which the user head faces. Then, feature extraction is carried out on the head portrait data of the user by adopting a feature extraction algorithm, position information of face contours, eyes, a nose and a mouth is reserved, extra details are filtered to obtain the feature data of the head portrait, a deep learning algorithm based on a Convolutional Neural Network (CNN) is used for training by utilizing a large amount of collected sample data, a certain training error threshold value is set, and a head angle judgment model which takes the head portrait of the user as an input parameter and the head angle of the user as an output parameter is obtained.
In one possible implementation, when the head is turned to different angles at the user position by using a model or a real person, the head images 504, 505 and 506 of the user shot by the cameras 501, 502 and 503 and the head angle of the user 507 are recorded, and such a large amount of sample data can be used as training data for obtaining the head angle judgment model, and a head angle error threshold value is set, and an optional threshold value is set to 5 degrees. When the head angle judgment model obtained through training is applied to an actual method for adjusting the balance of the binaural hearing aid, the judgment of the head angle of the user does not exceed 5 degrees, and the method has high precision and can meet the actual requirements of the user.
In the embodiment, considering that the fitting engineer needs to perform a large amount of hearing aid fitting work for a plurality of different users every day, and the balance adjustment of the binaural hearing aid is only a part of the fitting process, the invention provides a head angle determination model acquisition method, which is a basis for quickly acquiring the head angle of the user through a head portrait of the user shot by a camera.
In a possible implementation, only one camera for determining the head angle of the user may be installed on the top of the user or at another position where the head image of the user can be clearly captured, and before determining the head angle of the user based on the head image of the user captured by the one camera, the head angle determination model acquisition method described in the above embodiment may still be used to construct the feature data extracted by the image processing technique inputted to the head image of the user captured by the one camera, and output the head angle determination model of the head angle of the user corresponding to the head image of the user, for determining the head angle of the user in the balance adjustment of the hearing aid.
In the embodiment, the camera is used for acquiring the head portrait of the user and obtaining the head angle of the user through the image recognition mode (or other image recognition technologies of known technologies). In fact, however, other ways of acquiring the head angle of the user may be adopted, for example, a head-mounted head angle acquiring device with an angle sensor may be adopted to acquire the head angle of the user more accurately without considering implementation cost and efficiency.
In yet another embodiment of the present invention, the process of determining the target gain difference according to the angle deviation between the head angle of the user and the target sound source in the above steps of fig. 2 is described.
The angle of the user's head is determined by the user's head portrait photographed by the camera through 203 in the step of fig. 2, and the angle of the target sound emitting source is determined in advance, that is, is a known value. The invention discloses a method for determining a target gain difference according to an angle deviation between a head angle of a user and a target sound source.
In fig. 6 of the present embodiment, an alternative coordinate diagram with respect to the definition of angles is shown, which is applicable to the angle of the user's head and the angle of the target sound source. As shown in fig. 6, a polar coordinate system is established with the position of the user 601 as the o point, the o point as the pole, the ox as the polar axis, and the clockwise direction as the positive direction of the angle, and the following angles of the head of the user and the angle of the target sound source are measured in the coordinate system shown in fig. 6.
The angular deviation between the head angle of the user and the angle of the target sound source refers to an angular deviation from the head angle of the user to the target sound source, and is positive when the deviation is clockwise, and negative when the deviation is counterclockwise. In one possible implementation, the head angle of the user is
(120 degrees) with the angle of the target sound source being
(90 degrees), then the angular deviation is
(-30 degree).
The target gain difference reflects the adjustment amount of the hearing aid gain, the unit is db, the hearing defects of different types or degrees can be compensated by adjusting the gain of the hearing aid, and when the hearing aid has better compensation effect on the hearing defects, the smaller the angle deviation between the head angle of the user and the target sound source is in the fitting process, so that the angle deviation value between the head angle of the user and the target sound source has a certain relation with the target gain difference.
The invention discloses a relation between a user head angle and an angle deviation value x and a target gain difference g of a target sound source, which satisfies the following relation:
wherein x has a definition of [ - π, π ], where a positive value indicates a clockwise direction of the deviation angle and a negative value indicates a counterclockwise direction. The relation shows that when the absolute value of the angle deviation value of the head angle of the user and the target sound source is less than 45 degrees, the absolute value is in positive correlation with the target gain difference, when the absolute value is greater than or equal to 45 degrees, the target gain difference is 10, when the absolute value is less than or equal to-45 degrees, the target gain difference is-10, the range of the target gain difference is-10, and the unit is db.
In one possible implementation, as shown in FIG. 6, the
sound source 602 is a sound box at an angle of
The determination of the target gain difference in a set of possible occurrences is shown in table 1.
In the embodiment, the target gain difference is determined according to the angle deviation between the head angle of the user and the target sound source, so that a foundation is laid for adjusting the gain of the hearing aid and enabling the user to accurately distinguish the position of the target sound source.
Based on the embodiment corresponding to fig. 2, this embodiment describes a process of performing corresponding parameter adjustment on a hearing aid worn by a user according to a target gain difference to complete balance adjustment of a binaural hearing aid.
The dispenser may enable the hearing aid to give the hearing-impaired user different capabilities to distinguish between sounds, including the source of the sound, by adjusting the gain of the hearing aid. In the method disclosed by the present invention, the fitter only needs to adjust the gain of the hearing aid when the absolute value of the target gain difference exceeds the preset threshold range, which means that the current parameter setting of the hearing aid cannot meet the fitting target for the user to identify the sound source, that is, the hearing aid balance of the user is not adjusted, and further adjustment of the hearing aid gain is needed, and optionally, the threshold of the absolute value of the target gain difference can be preset to 1.
It will be appreciated that in an ideal situation the number of devices in the test sound source cluster corresponds to the maximum possible number of times the fitter needs to adjust the gain of the hearing aid, and in practice the number of sound boxes in the test sound box cluster is 7, indicating that the fitter needs to test the user with sound boxes at different angles, i.e. a total of 7 tests are required, in one possible implementation the absolute values of the target gain differences in the 1 st to 6 th tests exceed a preset threshold value 1, whereas in the 7 th test the absolute values of the target gain differences are within the threshold value 1, then the hearing aid needs to make gain parameter adjustments in the previous 6 tests, whereas the 7 th test does not require, in another possible implementation the absolute values of the target gain differences in the 1 st to 6 tests exceed the preset threshold value 1, and in the 7 th test, the absolute value of the target gain difference still exceeds the threshold 1, then the hearing aid needs gain parameter adjustments in 7 tests to complete the binaural hearing aid balance adjustment. Under other conditions, after the step of testing the balance adjustment corresponding to 1 sound box in the sound box group is completed, and after the step of testing the balance adjustment corresponding to other sound boxes in the sound box group, the sound box corresponding to the step of originally completing the balance adjustment may have the possibility of unbalance again, and then the step of balance adjustment needs to be performed again. In a word, the step of testing the balance adjustment of the sound boxes in the sound box group is performed for multiple times until all the sound boxes in the whole test sound box group are in a balanced state.
The invention discloses a method for guiding a dispenser to adjust the gain of a hearing aid by a target gain difference, namely a method for determining the adjustment amount of the gain of the hearing aid when the absolute value of the target gain difference is in a threshold range.
For convenience of explanation, the target gain difference is denoted by g, the preset threshold for the target gain difference is denoted by t, when g > t, the left ear gain of the hearing aid is increased by g/2, and the right ear gain of the hearing aid is decreased by g/2, when g < -t, the left ear gain of the hearing aid is decreased by g/2, and the right ear gain of the hearing aid is increased by g/2, and when g is in the range-t, the condition that the hearing aid does not need to be adjusted is explained. Wherein, the range of g is-10 to 10, the specific obtaining manner has been described in the previous embodiment, which is not described herein, and t is a positive number greater than 0.
In one possible implementation, t is preset to 1db and the confirmation of the adjustment of the hearing aid gain over a set of possible occurrences is shown in table 2.
Target gain difference g (db)
|
Left ear hearing aid gain adjustment (db)
|
Right ear hearing aid gain adjustment (db)
|
10
|
5
|
-5
|
1
|
Is not adjusted
|
Is not adjusted
|
-1
|
Is not adjusted
|
Is not adjusted
|
-10
|
-5
|
5 |
TABLE 2
As shown in table 2, when g is 10, it can be seen from the above embodiment that the angle deviation between the head angle of the user and the target sound source is not less than the angle deviation
(45 degrees), when g is 1, the angular deviation is about 0.08rad (4.5 degrees), no adjustment is needed for the hearing aid, and the parameters of the hearing aid can meet the requirement of the user for identifying the direction of the target sound emitting sourceAnd (4) precision.
In this embodiment, a method for determining an adjustment amount of a hearing aid gain in a binaural hearing aid balance adjustment is provided, and meanwhile, through repeated adjustments, a gain parameter of a hearing aid can meet more and more fitting targets for a fitter to complete the binaural hearing aid balance adjustment, so that an effective and feasible method is provided for the binaural hearing aid balance adjustment.
It should be added that the above embodiment describes the adjustment process of the angle deviation value x and the target gain difference g between the head angle of the user and the target sound source in a specific case, which is only used to show a specific implementation process of the embodiment, wherein the parameters of the relationship between the angle deviation value x and the target gain difference g between the head angle of the user and the target sound source and the preset threshold value can be adjusted and changed according to the fitting conditions and actual needs for balance adjustment, and the protection scope of the present invention should not be limited thereby.
The present embodiment will be described more fully in detail with reference to fig. 7, based on the embodiment shown in fig. 2, for the method of adjusting the balance of a binaural hearing aid disclosed in the present invention. Fig. 7 is a flow chart of a method for binaural hearing aid balance adjustment.
The sound source in this embodiment is selected as a sound box for explanation, and is the same as the embodiment corresponding to fig. 2.
In step 701, marking all the sound boxes as unbalanced states;
in step 702, randomly selecting a sound box in an unbalanced state to play sound;
in step 703, determining the head angle of the user through a head angle determination model according to the head images of the user shot by the cameras on the No. 1, No. 4 and No. 7 sound boxes;
in step 704, comparing the head angle of the user with the angle of the sounding sound box, and calculating an angle deviation;
step 705, determining a target gain difference according to the angle deviation;
in step 706, if the target gain difference is smaller than the preset threshold, go to step 708, otherwise go to step 707;
in step 707, according to the target gain difference, performing corresponding parameter adjustment on the hearing aid worn by the user;
in step 708, setting the target sound box to be in a balanced state;
in step 709, if there is a sound box in an unbalanced state, the process goes to step 702, otherwise, it indicates that the binaural hearing aid balance adjustment has been tested.
It should be noted that, in an ideal case, the adjustment effect of the hearing aid gain is not conflicting during the testing and adjustment of the hearing aid according to the unbalanced sound boxes at different angles.
In a possible implementation, the testing and matching engineer firstly selects the sound box No. 1 for testing, and adjusts the gain of the hearing aid according to the above steps, and the specific process is not repeated, so that when the sound box No. 1 is heard, the angle deviation between the angle of the position where the sound box No. 1 is located and the angle of the head of the user is within the set threshold range, then the sound box No. 1 is marked as a balanced state, and then the testing and matching engineer continuously selects the sound box group, optionally selects the sound box No. 3 for playing, and still adjusts the gain of the hearing aid according to the above steps, so that when the sound box No. 3 is heard, the angle deviation between the angle of the position where the sound box No. 3 is located and the angle of the head of the user is within the set threshold range, and meanwhile, the sound box No. 1 should still be in the.
The embodiment of the invention provides a method for balancing and adjusting a binaural hearing aid, which comprises the steps of firstly controlling a target sound source to make sound, judging the head angle of a user according to a head portrait of the user shot by a camera, adjusting gain parameters of the hearing aid according to an angle deviation between the angle of the target sound source and the head angle of the user until the angle deviation between the angle of the target sound source and the head angle of the user is smaller than a preset threshold value, marking the target sound source as a balanced state, then testing a next sound box in an unbalanced state until all the sound boxes are in the balanced state, so that the user can achieve binaural balance after wearing the hearing aid, the defect that the user binaural balance adjustment is not fully considered in an experimental and matching link is overcome, and more complete and effective experimental and matching service is provided for the user.
The following are embodiments of the apparatus of the present invention that may be used to perform embodiments of the method of the present invention. For details which are not disclosed in the embodiments of the apparatus of the present invention, reference is made to the embodiments of the method of the present invention.
Referring to fig. 8, a block diagram of a device for adjusting the balance of a binaural hearing aid according to an embodiment of the invention is shown. The device has the functions of executing the method examples, and the functions can be realized by hardware or corresponding software executed by the hardware. The apparatus may include:
a plurality of fixed sound sources;
a control playing module 801, configured to control a target sound source to play sound, where the target sound source is randomly selected;
a first determining module 802 for determining a head angle of a user;
a second determining module 803, which calculates an angle deviation according to the head angle of the user and the angle of the target sound source, and determines a target gain difference;
and an adjusting module 804, configured to perform corresponding parameter adjustment on the hearing aid worn by the user according to the target gain difference.
Optionally, the control playing module 801 is configured to:
the sound sources in the fitting condition form a test sound source group, the test sound source group is divided into a balance group and an unbalance group according to the balance state of each sound source in the test sound source group, wherein all the sound sources in the balance group are in the balance state, and all the sound sources in the unbalance group are in the unbalance state;
the target sound source is randomly selected from a group of imbalances in the test sound source population.
Optionally, the first determining module 802 is configured to:
acquiring corresponding head photos shot by the camera at the test position of the user when the head is at different angles;
and extracting feature data of the head photo, and establishing a head angle judgment model by combining the head angle corresponding to the head photo by adopting a data-driven modeling method, wherein the input parameter of the head angle judgment model is the feature data of the head photo, and the output parameter is the head angle.
Optionally, the second confirming module 803 is configured to:
when the absolute value of the angle deviation does not exceed a preset threshold range, the target gain difference and the angle deviation are in a positive correlation relationship;
when the absolute value of the angular deviation exceeds a preset threshold range, the value of the target gain difference is unchanged.
Optionally, the adjusting module 804 is configured to:
when the absolute value of the target gain difference exceeds a preset threshold range, the gain of the hearing aid needs to be adjusted.
In this embodiment, the host controls the target sound source to play sound, where the target sound source is randomly selected from the test sound source group by the host, and then the user can face the face to the position of the target sound source under the instruction of the fitter after hearing the sound played by the target sound source, and then the target gain difference is determined according to the angle deviation between the head angle of the user and the target sound source, and finally, according to the target gain difference, corresponding parameter adjustment is performed on the hearing aid worn by the user, so as to complete binaural hearing aid balance adjustment. Through the mode, the testing and matching technician can efficiently and conveniently guide the user with hearing impairment or hearing deficiency to complete the balance adjustment test of the binaural hearing aid, the aim of balance adjustment of the binaural hearing aid is fulfilled, and the user can more accurately distinguish the source of sound in daily work and life after wearing the adjusted hearing aid.
Furthermore, the process that the testing and matching engineer controls the target loudspeaker boxes to play sound through the host is added, the testing loudspeaker box group is uniformly controlled through the host console, the testing progress can be conveniently managed through the visual interface, and for example, the testing and matching engineer can clearly know the quantity of the loudspeaker boxes in a balanced state according to the background shape of the corresponding entity loudspeaker boxes on the interface, namely the quantity of the loudspeaker boxes in the balanced state.
Considering that the fitter is performing a lot of hearing aid fitting work for many different users each day, the binaural hearing aid balance adjustment is only part of the fitting process. The invention provides a method for acquiring a head angle judgment model, which is the basis for acquiring the head angle of a user through a head portrait of the user shot by a camera.
And determining a target gain difference according to the angle deviation between the head angle of the user and the target sound source, and laying a foundation for adjusting the gain of the hearing aid and enabling the user to accurately distinguish the position of the target sound source.
The method has the advantages that in the balance adjustment of the binaural hearing aid, the gain parameters of the hearing aid can meet the fitting target of the fitting operator to finish the balance adjustment of the binaural hearing aid more and more through repeated adjustment, and the balance adjustment of the binaural hearing aid can be effectively realized.
Fig. 9 is a schematic diagram showing a structure of a host according to an exemplary embodiment of the present invention. The host computer includes: a processor 901, a memory 902, and a bus 903.
The processor 901 comprises one or more processing cores, the processor 902 is connected to the processor 901 via a bus 903, the memory 903 is used for storing program instructions, and the processor 901 realizes the method for binaural hearing aid balance adjustment provided by the above-described embodiments when executing the program instructions in the memory 902.
Alternatively, the memory 902 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The present invention also provides a computer readable storage medium having stored therein at least one instruction, at least one program, code set or set of instructions, which is loaded and executed by the processor to implement the method of binaural hearing aid balance adjustment provided by the above method embodiments.
Alternatively, the present invention also provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of binaural hearing aid balance adjustment according to the above aspects.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by hardware related to instructions of a program, where the program may be stored in a computer readable storage medium, and the above mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.