CN113259825B - Low-complexity amplitude-frequency response later-period correction method for digital hearing aid - Google Patents

Low-complexity amplitude-frequency response later-period correction method for digital hearing aid Download PDF

Info

Publication number
CN113259825B
CN113259825B CN202110618842.3A CN202110618842A CN113259825B CN 113259825 B CN113259825 B CN 113259825B CN 202110618842 A CN202110618842 A CN 202110618842A CN 113259825 B CN113259825 B CN 113259825B
Authority
CN
China
Prior art keywords
amplitude
frequency response
sub
band
correction
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
Application number
CN202110618842.3A
Other languages
Chinese (zh)
Other versions
CN113259825A (en
Inventor
王成超
孟宪军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Tianyue Electronic Technology Co ltd
Original Assignee
Nanjing Tianyue Electronic Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Tianyue Electronic Technology Co ltd filed Critical Nanjing Tianyue Electronic Technology Co ltd
Priority to CN202110618842.3A priority Critical patent/CN113259825B/en
Publication of CN113259825A publication Critical patent/CN113259825A/en
Application granted granted Critical
Publication of CN113259825B publication Critical patent/CN113259825B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
    • H04R25/50—Customised settings for obtaining desired overall acoustical characteristics
    • H04R25/505—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
    • H04R3/00—Circuits for transducers, loudspeakers or microphones
    • H04R3/04—Circuits for transducers, loudspeakers or microphones for correcting frequency response

Abstract

The invention discloses a later-stage correction method of low-complexity amplitude-frequency response for a digital hearing aid, which comprises the steps of firstly removing direct current offset of an original signal through a high-pass filter, obtaining a plurality of sub-bands through a sub-band divider composed of a filter bank, sending amplitude-frequency response to be corrected of the sub-band signals obtained through division and expected amplitude-frequency response of the digital hearing aid into a later-stage correction system for later-stage correction of the amplitude-frequency response of the digital hearing aid, adaptively selecting a proper amplitude-frequency response correction method combination strategy and correction parameters through an amplitude-frequency response correction control unit by the system, finally obtaining corrected amplitude-frequency response approaching to an expected value, facilitating simple and efficient configuration of the amplitude-frequency response of the digital hearing aid, being capable of adaptively performing later-stage correction on the amplitude-frequency response, avoiding calculation of FFT and IFFT and greatly reducing the complexity of an algorithm, the method is more suitable for the design of digital hearing aids, and has good application prospect.

Description

Low-complexity amplitude-frequency response later-period correction method for digital hearing aid
Technical Field
The invention relates to the technical field of signal processing, in particular to a low-complexity amplitude-frequency response later-stage correction method for a digital hearing aid.
Background
With the rapid development of industry and the aggravation of aging phenomenon, the number of people suffering from various degrees of hearing disorder diseases is continuously increasing. The method for improving the hearing of a hearing-impaired patient by selecting a proper hearing aid is the most convenient, quick and common means in the existing medical level, compared with the defects that the self-adaptive capacity of a simulation hearing aid to various complex environments is poor, the discomfort of the hearing-impaired patient is easily caused by overlarge output, even secondary damage is caused, and the like, the digital hearing aid has the advantages of higher sound precision, smaller distortion, longer service life, programmable realization of more complex operation, stronger adaptability to different complex environments and the like, so that the digital hearing aid has wide development space in the field of hearing aids, and further influences the research on the algorithm of the digital hearing aid.
Currently, digital hearing aid algorithms mainly include key algorithms such as loudness compensation, echo cancellation, noise suppression, and the like. Because the hearing of the hearing-impaired person and the common person is greatly different in different frequencies, the loudness compensation algorithm becomes the most basic and important technology in the digital hearing aid. Moreover, even if the amplitude-frequency response is corrected in each sub-band completely according to the hearing threshold diagram of the hearing impaired person, the fixed amplitude-frequency response of the hearing impaired person cannot meet the use requirements of different hearing impaired persons due to the difference between the hearing loss degree of each hearing impaired person and the individual preference, and therefore, the amplitude-frequency response correction according to the hearing threshold diagram and the expected frequency response of the hearing impaired person is not only necessary, but also has an extremely important role in customizing and improving the hearing experience of the hearing impaired person.
In terms of frequency response correction methods, researchers have proposed many feasible methods: the disclosed CN1870135A patent CN1870135A discloses a method for compensating frequency response of a digital hearing aid based on masking curve, which implements frequency response compensation of the digital hearing aid by using methods of calculating masking threshold and dividing critical band, but includes calculation processes of FFT, IFFT, calculating spread critical band spectrum, calculating masking threshold, etc., so that it is not suitable for digital hearing aid application scenarios with low complexity, low computation amount and low power consumption. Therefore, how to overcome the above problems is a problem to be solved currently.
Disclosure of Invention
It is an object of the present invention to overcome the problems of the prior art digital hearing aid algorithms. The low-complexity amplitude-frequency response later-stage correction method for the digital hearing aid can adaptively perform later-stage correction on the amplitude-frequency response, avoids calculation of FFT (fast Fourier transform) and IFFT (inverse fast Fourier transform), greatly reduces the complexity of the algorithm, can dynamically adjust parameters of each filter in the algorithm, greatly shortens the fitting and development periods of hearing aid fitters and digital hearing aid developers, more efficiently configures the amplitude-frequency response of the digital hearing aid, is more suitable for the design of the digital hearing aid, and has good application prospect.
In order to achieve the purpose, the invention adopts the technical scheme that:
a late stage correction method for low complexity amplitude frequency response of digital hearing aid comprises the following steps,
step (A), removing direct current offset of an original signal through a high-pass filter, and obtaining a plurality of sub-band signals through segmentation through a sub-band splitter consisting of a filter bank;
step (B), scanning frequency of the plurality of sub-band signals obtained by division one by one to obtain amplitude-frequency response signals to be corrected corresponding to the plurality of sub-band signals;
step (C), sending the amplitude-frequency response signals to be corrected and the expected target amplitude-frequency response signals corresponding to the plurality of sub-band signals into a later-stage correction unit system together for carrying out later-stage correction on the amplitude-frequency response of the digital hearing aid;
and (D) adaptively selecting a proper amplitude-frequency response correction method combination strategy and correction parameters through a later correction unit system to obtain corrected amplitude-frequency response signals which are corresponding to a plurality of sub-band signals and approximate to expected values and correspond to amplitude-frequency response signals to be corrected.
The low-complexity amplitude-frequency response post-correction method for the digital hearing aid comprises the following steps of (A) removing direct current offset of an original signal through a high-pass filter, and obtaining a plurality of sub-band signals through division by a sub-band divider formed by a filter bank,
(A1) removing DC offset from the original signal by a high-pass filter to obtain a digital audio signal x with frames and DC offset removedk；
(A2) A digital audio signal xkSending the signals into an IIR sub-band segmentation filter bank to obtain M sub-band signals xk(i, m), where k is the number of frames, i is the number of samples, and m is the number of subbands.
In the aforementioned method for late modification of low-complexity amplitude-frequency response for digital hearing aids, the value of M is 4.
In the method for late-stage correction of low-complexity amplitude-frequency response for a digital hearing aid, the frequency of the divided sub-band signals is swept one by one to obtain amplitude-frequency response signals to be corrected corresponding to the sub-band signals, and the specific process is as follows: using the M subband signals x obtained in the swept frequency signal pair (A2)kAnd (i, M) scanning frequency one by one to obtain M corresponding amplitude-frequency responses r (M).
The low-complexity amplitude-frequency response post-correction method for the digital hearing aid, which is used for the step (D), adaptively selecting a proper amplitude-frequency response correction method combination strategy and correction parameters through a post-correction unit system to obtain a corrected amplitude-frequency response signal which is close to an expected value and corresponds to amplitude-frequency response signals to be corrected and corresponds to a plurality of sub-band signals, specifically comprises the following steps,
(D1) comparing the amplitude-frequency response correction control unit in the later correction unit system with the amplitude-frequency response signal to be corrected corresponding to the input sub-band signals and the expected target amplitude-frequency response signal to obtain the error vector E ═ E of each sub-band1,e2,e3,e4]T；
(D2) The amplitude-frequency response correction control unit corrects the error E of each sub-band according to the error E of each sub-band in the error vector E of each sub-bandmAnd generating a sub-band gating vector K ═ K in relation to an error threshold constant th1,k2,k3,k4]TAnd converting the sub-band gating vector to generate a sub-band correction coefficient vector K '═ K'1,k′2,k′3,k′4]TSetting a sub-band to be corrected and a correction coefficient thereof;
(D3) generating a filter gating vector C ═ C by an amplitude-frequency response correction control unit1,c2,c3,c4]Setting different correction methods;
(D4) and (3) sending the output correction matrix A which is K' C to the digital hearing aid for configuration through the amplitude-frequency response correction control unit, and performing frequency sweeping operation in the step (B) again after the digital hearing aid runs off line to obtain corrected amplitude-frequency response signals which are corresponding to a plurality of sub-band signals and approximate to expected values and correspond to amplitude-frequency response signals to be corrected.
The method for late correction of low-complexity amplitude-frequency response for digital hearing aid utilizes the signals x to M sub-band signals obtained from the frequency sweep signal pair (A2)kAnd (i, M) carrying out frequency sweeping one by one to obtain M corresponding amplitude-frequency responses r (M), wherein the frequency range of the frequency sweeping signal is 100-8000 Hz.
The method for late-stage correction of low-complexity amplitude-frequency response for digital hearing aids, (D2), error E for each subbandmAs shown in the following formula,
Figure GDA0003236995600000041
wherein n ismThe number of equally spaced frequency points in the mth sub-band is 10 Hz; q is a transition zone constant, and q is taken to be 3; r (j) is expected amplitude-frequency response, r (j) is amplitude-frequency response to be corrected, and j is a frequency point serial number;
sub-band gating vector K ═ K1,k2,k3,k4]TEach element k1，k2，k3，k4The value taking method comprises the following steps: if | EmH is less than or equal to | and represents that the error is within an acceptable range, the subband is considered to be not required to be corrected, and k is takenm0, otherwise km＝EmTh is an error threshold constant, and 0.02 × r (m) is taken as th, wherein r (m) is an expected amplitude-frequency response of the mth subband;
sub-band gating vector K ═ K1,k2,k3,k4]TTransform generates sub-band correction coefficient vector K '═ K'1,k′2,k′3,k′4]THere, the
Figure GDA0003236995600000051
In the aforementioned late stage correction method for low-complexity amplitude-frequency response of digital hearing aid, (D3), the amplitude-frequency response correction control unit generates filter gating vector C ═ C1,c2,c3,c4]Setting different correction methods, wherein the gating vector C of the filter is ═ C1,c2,c3,c4]Wherein each element takes the value of 0 or 1, c1Corresponding to the first filter, c2Corresponding to the second filter, c3Corresponding to the second filter, c4Corresponding to the fourth filter.
In the aforementioned late stage correction method for low-complexity amplitude-frequency response of a digital hearing aid, the first filter is an integrator; the second filter is a first order differential differentiator; the third filter is a central differential differentiator; the fourth filter is a narrow-band differentiator.
In the aforementioned late stage correction method for low-complexity amplitude-frequency response of digital hearing aid, (D3), the amplitude-frequency response correction control unit generates filter gating vector C ═ C1,c2,c3,c4]Setting different correction methods, specifically comprising the following steps,
(D301) if e is1If the amplitude-frequency response of the first sub-band needs to be improved, the step (D311) is carried out; otherwise, entering the step (D302);
(D302) if e is1Is < -th and e2＞th，e3＞th，e4If the amplitude-frequency response of the first sub-band needs to be reduced and the amplitude-frequency responses of the second, third and fourth sub-bands need to be improved, the step (D312) is carried out; otherwise, entering the step (D303);
(D303) if e is1＜-th，e4Is < -th and e2≤th，e3Th is less than or equal to th, which indicates that the amplitude-frequency response of the first sub-band and the amplitude-frequency response of the fourth sub-band need to be reduced but the amplitude-frequency response of the second sub-band and the amplitude-frequency response of the third sub-band need not to be reduced, then step (D313) is carried out; otherwise, entering the step (D304);
(D304) if e is1Is < -th and e2＞th，e3＞th，e4If the amplitude-frequency response of the first sub-band needs to be reduced, the amplitude-frequency response of the second sub-band and the amplitude-frequency response of the third sub-band are amplified, and the amplitude-frequency response of the fourth sub-band is not amplified, the step (D314) is carried out;
(D311) said subband correction coefficient vector
Figure GDA0003236995600000061
Filter gating vector C ═ 1,0,0,0]；
(D312) Said subband correction coefficient vector
Figure GDA0003236995600000062
Filter gating vector C ═ 0,1,0]；
(D313) Said subband correction coefficient vector
Figure GDA0003236995600000063
Filter gating vector C ═ 0,0,1,0]；
(D314) Said subband correction coefficient vector
Figure GDA0003236995600000064
Filter gating vector C ═ 0,0,0,1]。
The invention has the beneficial effects that: the low-complexity amplitude-frequency response later-stage correction method for the digital hearing aid can adaptively perform later-stage correction on the amplitude-frequency response, avoids calculation of FFT (fast Fourier transform) and IFFT (inverse fast Fourier transform), greatly reduces the complexity of the algorithm, can dynamically adjust parameters of each filter in the algorithm, greatly shortens the fitting and development periods of hearing aid fitters and digital hearing aid developers, more efficiently configures the amplitude-frequency response of the digital hearing aid, is more suitable for the design of the digital hearing aid, and has good application prospect.
Drawings
Fig. 1 is a flow chart of the method for late-stage correction of low-complexity amplitude-frequency response for digital hearing aids according to the present invention;
fig. 2 is a diagram illustrating the effect of amplitude-frequency response correction according to an embodiment of the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
As shown in FIG. 1, the method for late stage correction of low complexity amplitude-frequency response for digital hearing aid of the present invention comprises the steps of firstly removing DC offset from an original signal by a high pass filter, obtaining a plurality of sub-bands by a sub-band splitter composed of a filter bank, sending the amplitude-frequency response to be corrected of the obtained sub-band signals and the expected amplitude-frequency response of the digital hearing aid into a late stage correction unit system for late stage correction of the amplitude-frequency response of the digital hearing aid, adaptively selecting a proper amplitude-frequency response correction method combination strategy and correction parameters by an amplitude-frequency response correction control unit, and finally obtaining a corrected amplitude-frequency response approaching to an expected value, wherein the algorithm has low complexity, is convenient for simply and efficiently configuring the amplitude-frequency response of the digital hearing aid, and specifically comprises the following steps,
step (A), removing direct current offset from an original signal by a high-pass filter, and obtaining a plurality of sub-band signals by dividing the original signal by a sub-band divider composed of a filter bank,
(A1) removing DC offset from the original signal by a high-pass filter to obtain a digital audio signal x with frames and DC offset removedk；
(A2) A digital audio signal xkSending the signals into an IIR sub-band segmentation filter bank to obtain M sub-band signals xk(i, m), wherein k is the frame number, i is the number of sample points, and m is the number of sub-bands;
preferably, the number of divided subbands M is 4, and each subband frequency range is as shown in table 1:
TABLE 1 frequency ranges of the sub-bands
Figure GDA0003236995600000071
Step (B), scanning frequency one by one for the sub-band signals obtained by division to obtain amplitude-frequency response signals to be corrected corresponding to the sub-band signals, specifically M sub-band signals x obtained in the scanning frequency signal pair (A2)k(i, M) carrying out frequency sweeping one by one to obtain M corresponding amplitude-frequency responses r (M), wherein the frequency range of the frequency sweeping signal is 100-8000 Hz;
step (C), sending the amplitude-frequency response signals to be corrected and the expected target amplitude-frequency response signals corresponding to the plurality of sub-band signals into a later-stage correction unit system together for carrying out later-stage correction on the amplitude-frequency response of the digital hearing aid;
step (D), a proper amplitude-frequency response correction method combination strategy and correction parameters are selected in a self-adaptive manner through a later-stage correction unit system, and corrected amplitude-frequency response signals which are corresponding to a plurality of sub-band signals and approximate to expected values and correspond to amplitude-frequency response signals to be corrected are obtained,
(D1) modifying the input signals by the amplitude-frequency response modification control unit in the post-modification unit systemComparing the amplitude-frequency response signal to be corrected corresponding to the band signal with the expected target amplitude-frequency response signal to obtain an error vector E ═ E of each sub-band1,e2,e3,e4]T；
(D2) The amplitude-frequency response correction control unit corrects the error E of each sub-band according to the error E of each sub-band in the error vector E of each sub-bandmAnd generating a sub-band gating vector K ═ K in relation to an error threshold constant th1,k2,k3,k4]TAnd converting the sub-band gating vector to generate a sub-band correction coefficient vector K '═ K'1,k′2,k′3,k′4]TAnd setting the sub-band to be corrected and the correction coefficient thereof, wherein the error E of each sub-bandmAs shown in the following formula,
Figure GDA0003236995600000081
wherein n ismThe number of equally spaced frequency points in the mth sub-band is 10 Hz; q is a transition zone constant, and q is taken to be 3; r (j) is expected amplitude-frequency response, r (j) is amplitude-frequency response to be corrected, and j is a frequency point serial number;
sub-band gating vector K ═ K1,k2,k3,k4]TEach element k1，k2，k3，k4The value taking method comprises the following steps: if | EmH is less than or equal to | and represents that the error is within an acceptable range, the subband is considered to be not required to be corrected, and k is takenm0, otherwise km＝EmTh is an error threshold constant, and 0.02 × r (m) is taken as th, wherein r (m) is an expected amplitude-frequency response of the mth subband;
sub-band gating vector K ═ K1,k2,k3,k4]TTransform generates sub-band correction coefficient vector K '═ K'1,k′2,k′3,k′4]THere, the
Figure GDA0003236995600000082
(D3) Generating a filter gating vector C ═ C by an amplitude-frequency response correction control unit1,c2,c3,c4]Setting different correction methods, wherein the gating vector C of the filter is ═ C1,c2,c3,c4]Wherein each element takes the value of 0 or 1, c1Corresponding to the first filter, c2Corresponding to the second filter, c3Corresponding to the second filter, c4Corresponding to a fourth filter, the first filter is an integrator; the second filter is a first order differential differentiator; the third filter is a central differential differentiator; the fourth filter is a narrow-band differentiator, and the different modification methods are specifically,
(D301) if e is1If the amplitude-frequency response of the first sub-band needs to be improved, the step (D311) is carried out; otherwise, entering the step (D302);
(D302) if e is1Is < -th and e2＞th，e3＞th，e4If the amplitude-frequency response of the first sub-band needs to be reduced and the amplitude-frequency responses of the second, third and fourth sub-bands need to be improved, the step (D312) is carried out; otherwise, entering the step (D303);
(D303) if e is1＜-th，e4Is < -th and e2≤th，e3Th is less than or equal to th, which indicates that the amplitude-frequency response of the first sub-band and the amplitude-frequency response of the fourth sub-band need to be reduced but the amplitude-frequency response of the second sub-band and the amplitude-frequency response of the third sub-band need not to be reduced, then step (D313) is carried out; otherwise, entering the step (D304);
(D304) if e is1Is < -th and e2＞th，e3＞th，e4If the amplitude-frequency response of the first sub-band needs to be reduced, the amplitude-frequency response of the second sub-band and the amplitude-frequency response of the third sub-band are amplified, and the amplitude-frequency response of the fourth sub-band is not amplified, the step (D314) is carried out;
(D311) said subband correction coefficient vector
Figure GDA0003236995600000091
Filter gating vector C ═ 1,0,0,0]；
(D312) Said subband correction coefficient vector
Figure GDA0003236995600000092
Filter gating vector C ═ 0,1,0]；
(D313) Said subband correction coefficient vector
Figure GDA0003236995600000093
Filter gating vector C ═ 0,0,1,0]；
(D314) Said subband correction coefficient vector
Figure GDA0003236995600000101
Filter gating vector C ═ 0,0,0,1]；
(D4) And (3) sending the output correction matrix A which is K' C to the digital hearing aid for configuration through the amplitude-frequency response correction control unit, and performing frequency sweeping operation in the step (B) again after the digital hearing aid runs off line to obtain corrected amplitude-frequency response signals which are corresponding to a plurality of sub-band signals and approximate to expected values and correspond to amplitude-frequency response signals to be corrected.
According to the method for late stage correction of low-complexity amplitude-frequency response for digital hearing aid, in one embodiment, when the condition of step (D313) is met, step (D314) is carried out, and the selected sub-band correction coefficient vector is entered
Figure GDA0003236995600000102
Filter gating vector C ═ 0,0,0,1]The amplitude-frequency response later-period correction unit system sends the correction matrix A to K' C into the digital hearing aid through the serial communication interface, the hearing aid obtains the corrected amplitude-frequency response after sweeping frequency again, the amplitude-frequency response correction effect, as shown in fig. 2, the correction matrix can be observed on an amplitude-frequency response curve, on one hand, the amplitude-frequency response of the second sub-band is improved, and most of speaking sounds are concentrated in the middle frequency band, so that the voices are clearer, the improvement of speech intelligibility is facilitated, and on the other hand, the amplitude-frequency response of the first sub-band is reduced, so that the noise of the low frequency band is restrained to a certain extent.
In conclusion, the digital hearing aid-oriented low-complexity amplitude-frequency response post-correction method can adaptively perform post-correction on the amplitude-frequency response, avoids calculation of FFT (fast Fourier transform) and IFFT (inverse fast Fourier transform), greatly reduces the complexity of the algorithm, can dynamically adjust parameters of each filter in the algorithm, greatly shortens the fitting and development periods of hearing aid fitters and digital hearing aid developers, more efficiently configures the amplitude-frequency response of the digital hearing aid, is more suitable for the design of the digital hearing aid, and has a good application prospect.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (7)

1. A late stage correction method for low-complexity amplitude-frequency response of a digital hearing aid is characterized by comprising the following steps: comprises the following steps of (a) carrying out,
step (A), removing direct current offset of an original signal through a high-pass filter, and obtaining a plurality of sub-band signals through segmentation through a sub-band splitter consisting of a filter bank;
step (B), scanning frequency of the plurality of sub-band signals obtained by division one by one to obtain amplitude-frequency response signals to be corrected corresponding to the plurality of sub-band signals;
step (C), sending the amplitude-frequency response signals to be corrected and the expected target amplitude-frequency response signals corresponding to the plurality of sub-band signals into a later-stage correction unit system together for carrying out later-stage correction on the amplitude-frequency response of the digital hearing aid;
step (D), a proper amplitude-frequency response correction method combination strategy and correction parameters are selected in a self-adaptive manner through a later-stage correction unit system, and corrected amplitude-frequency response signals which are corresponding to a plurality of sub-band signals and approximate to expected values and correspond to amplitude-frequency response signals to be corrected are obtained,
(D1) through a later stageAn amplitude-frequency response correction control unit in the correction unit system compares amplitude-frequency response signals to be corrected corresponding to a plurality of input sub-band signals with an expected target amplitude-frequency response signal to obtain an error vector E ═ E of each sub-band1，e2，e3，e4]T；
(D2) The amplitude-frequency response correction control unit corrects the error E of each sub-band according to the error E of each sub-band in the error vector E of each sub-bandmAnd generating a sub-band gating vector K ═ K in relation to an error threshold constant th1，k2，k3，k4]TAnd converting the sub-band gating vector to generate a sub-band correction coefficient vector K '═ K'1，k′2，k′3，k′4]TSetting a sub-band to be corrected and a correction coefficient thereof;
(D3) generating a filter gating vector C ═ C by an amplitude-frequency response correction control unit1，c2，c3，c4]Setting different correction methods;
(D4) sending the output correction matrix A which is K' C to the digital hearing aid for configuration through an amplitude-frequency response correction control unit, performing frequency sweeping operation in the step (B) again after the digital hearing aid runs off line to obtain corrected amplitude-frequency response signals which are corresponding to a plurality of sub-band signals and approximate to expected values and correspond to amplitude-frequency response signals to be corrected,
therein (D2), error E of each sub-bandmAs shown in the following formula,
Figure FDA0003236995590000021
wherein n ismThe number of equally spaced frequency points in the mth sub-band is 10 Hz; q is a transition zone constant, and q is taken to be 3; r (j) is expected amplitude-frequency response, r (j) is amplitude-frequency response to be corrected, and j is a frequency point serial number;
sub-band gating vector K ═ K1,k2,k3,k4]TEach element k1，k2，k3，k4The value taking method comprises the following steps: if | EmH is less than or equal to | and represents that the error is within an acceptable range, the subband is considered to be not required to be corrected, and k is takenm0, otherwise km＝EmTh is an error threshold constant, and 0.02 × r (m) is taken as th, wherein r (m) is an expected amplitude-frequency response of the mth subband;
sub-band gating vector K ═ K1,k2,k3,k4]TTransform generates sub-band correction coefficient vector K '═ K'1,k′2,k′3,k′4]THere, the
Figure FDA0003236995590000022
(D3) Generating a filter gating vector C ═ C by an amplitude-frequency response correction control unit1,c2,c3,c4]Setting different correction methods, wherein the gating vector C of the filter is ═ C1,c2,c3,c4]Wherein each element takes the value of 0 or 1, c1Corresponding to the first filter, c2Corresponding to the second filter, c3Corresponding to the second filter, c4Corresponding to the fourth filter.
2. The method for late-stage modification of low-complexity amplitude-frequency response for digital hearing aids according to claim 1, wherein: step (A), removing direct current offset from an original signal by a high-pass filter, and obtaining a plurality of sub-band signals by dividing the original signal by a sub-band divider composed of a filter bank,
(A1) removing DC offset from the original signal by a high-pass filter to obtain a digital audio signal x with frames and DC offset removedk；
(A2) A digital audio signal xkSending the signals into an IIR sub-band segmentation filter bank to obtain M sub-band signals xk(i, m), where k is the number of frames, i is the number of samples, and m is the number of subbands.
3. The method for late-stage correction of low-complexity amplitude-frequency response for digital hearing aids according to claim 2, wherein: the value of M is 4.
4. The method for late-stage correction of low-complexity amplitude-frequency response for digital hearing aids according to claim 3, wherein: scanning frequency one by one for a plurality of sub-band signals obtained by division to obtain amplitude-frequency response signals to be corrected corresponding to the plurality of sub-band signals, and the specific process is as follows: using the M subband signals x obtained in the swept frequency signal pair (A2)kAnd (i, M) scanning frequency one by one to obtain M corresponding amplitude-frequency responses r (M).
5. The method for late-stage correction of low-complexity amplitude-frequency response for digital hearing aids according to claim 2, wherein: using the M subband signals x obtained in the swept frequency signal pair (A2)kAnd (i, M) carrying out frequency sweeping one by one to obtain M corresponding amplitude-frequency responses r (M), wherein the frequency range of the frequency sweeping signal is 100-8000 Hz.
6. The method for late-stage modification of low-complexity amplitude-frequency response for digital hearing aids according to claim 1, wherein: the first filter is an integrator; the second filter is a first order differential differentiator; the third filter is a central differential differentiator; the fourth filter is a narrow-band differentiator.
7. The method for late-stage modification of low-complexity amplitude-frequency response for digital hearing aids according to claim 1, wherein: (D3) generating a filter gating vector C ═ C by an amplitude-frequency response correction control unit1,c2,c3,c4]Setting different correction methods, specifically comprising the following steps,
(D301) if e is1If the amplitude-frequency response of the first sub-band needs to be improved, the step (D311) is carried out; otherwise, entering the step (D302);
(D302) if e is1Is < -th and e2＞th，e3＞th，e4＞thIf it is indicated that the amplitude-frequency response of the first sub-band needs to be reduced and the amplitude-frequency responses of the second, third and fourth sub-bands need to be improved, the procedure proceeds to step (D312); otherwise, entering the step (D303);
(D303) if e is1＜-th，e4Is < -th and e2≤th，e3Th is less than or equal to th, which indicates that the amplitude-frequency response of the first sub-band and the amplitude-frequency response of the fourth sub-band need to be reduced but the amplitude-frequency response of the second sub-band and the amplitude-frequency response of the third sub-band need not to be reduced, then step (D313) is carried out; otherwise, entering the step (D304);
(D304) if e is1Is < -th and e2＞th，e3＞th，e4If the amplitude-frequency response of the first sub-band needs to be reduced, the amplitude-frequency response of the second sub-band and the amplitude-frequency response of the third sub-band are amplified, and the amplitude-frequency response of the fourth sub-band is not amplified, the step (D314) is carried out;
(D311) said subband correction coefficient vector
Figure FDA0003236995590000041
Filter gating vector C ═ 1,0,0,0]；
(D312) Said subband correction coefficient vector
Figure FDA0003236995590000042
Filter gating vector C ═ 0,1,0]；
(D313) Said subband correction coefficient vector
Figure FDA0003236995590000043
Filter gating vector C ═ 0,0,1,0]；
(D314) Said subband correction coefficient vector
Figure FDA0003236995590000044
Filter gating vector C ═ 0,0,0,1]。
CN202110618842.3A 2021-06-03 2021-06-03 Low-complexity amplitude-frequency response later-period correction method for digital hearing aid Active CN113259825B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110618842.3A CN113259825B (en) 2021-06-03 2021-06-03 Low-complexity amplitude-frequency response later-period correction method for digital hearing aid

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110618842.3A CN113259825B (en) 2021-06-03 2021-06-03 Low-complexity amplitude-frequency response later-period correction method for digital hearing aid

Publications (2)

Publication Number Publication Date
CN113259825A CN113259825A (en) 2021-08-13
CN113259825B true CN113259825B (en) 2021-10-08

Family

ID=77186277

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110618842.3A Active CN113259825B (en) 2021-06-03 2021-06-03 Low-complexity amplitude-frequency response later-period correction method for digital hearing aid

Country Status (1)

Country Link
CN (1) CN113259825B (en)

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DK1869948T3 (en) * 2005-03-29 2016-05-02 Gn Resound As Hearing aid with adaptive compressor time constants
DE102005020317B4 (en) * 2005-05-02 2009-04-02 Siemens Audiologische Technik Gmbh Automatic gain adjustment on a hearing aid
US8566086B2 (en) * 2005-06-28 2013-10-22 Qnx Software Systems Limited System for adaptive enhancement of speech signals
US8005246B2 (en) * 2007-10-23 2011-08-23 Swat/Acr Portfolio Llc Hearing aid apparatus
CN112151044A (en) * 2020-09-23 2020-12-29 北京百瑞互联技术有限公司 Method, device and storage medium for automatically adjusting frequency response curve of Bluetooth playing device in LC3 audio encoder

Also Published As

Publication number Publication date
CN113259825A (en) 2021-08-13

Similar Documents

Publication Publication Date Title
CN103871421B (en) A kind of self-adaptation noise reduction method and system based on subband noise analysis
AU771444B2 (en) Noise reduction apparatus and method
CN101901602B (en) Method for reducing noise by using hearing threshold of impaired hearing
CN109493877B (en) Voice enhancement method and device of hearing aid device
CN105706466B (en) Hearing aid with probabilistic hearing compensation
CN111261182B (en) Wind noise suppression method and system suitable for cochlear implant
CN105741849A (en) Voice enhancement method for fusing phase estimation and human ear hearing characteristics in digital hearing aid
CN102543095B (en) For reducing the method and apparatus of the tone artifacts in audio processing algorithms
US20030223597A1 (en) Adapative noise compensation for dynamic signal enhancement
CN106788471B (en) A kind of sound spectrum control system and method for noise circumstance test
CN113993053B (en) Channel self-adaptive digital hearing aid wide dynamic range compression method
Chiea et al. New insights on the optimality of parameterized Wiener filters for speech enhancement applications
TWI819478B (en) Hearing device with end-to-end neural network and audio processing method
CN117321681A (en) Speech optimization in noisy environments
EP2368243A1 (en) Systems and methods for improving the intelligibility of speech in a noisy environment
CN113259825B (en) Low-complexity amplitude-frequency response later-period correction method for digital hearing aid
CN110708651B (en) Hearing aid squeal detection and suppression method and device based on segmented trapped wave
CN116168719A (en) Sound gain adjusting method and system based on context analysis
CN109874088A (en) A kind of method and apparatus adjusting sound pressure level
CN113257268B (en) Noise reduction and single-frequency interference suppression method combining frequency tracking and frequency spectrum correction
US20230169987A1 (en) Reduced-bandwidth speech enhancement with bandwidth extension
US20060087380A1 (en) Method for limiting the dynamic range of audio signals, and circuit arrangement for this purpose
WO2022150286A1 (en) Determining dialog quality metrics of a mixed audio signal
US11445307B2 (en) Personal communication device as a hearing aid with real-time interactive user interface
WO2021055513A1 (en) Mitigating acoustic feedback in hearing aids with frequency warping by all-pass networks

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