CN114584909B - Digital hearing aid howling suppression system and suppression method thereof - Google Patents
Digital hearing aid howling suppression system and suppression method thereof Download PDFInfo
- Publication number
- CN114584909B CN114584909B CN202210463650.4A CN202210463650A CN114584909B CN 114584909 B CN114584909 B CN 114584909B CN 202210463650 A CN202210463650 A CN 202210463650A CN 114584909 B CN114584909 B CN 114584909B
- Authority
- CN
- China
- Prior art keywords
- signal
- hearing aid
- filter
- far
- adaptive filter
- 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
-
- 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
- H04R25/45—Prevention of acoustic reaction, i.e. acoustic oscillatory feedback
- H04R25/453—Prevention of acoustic reaction, i.e. acoustic oscillatory feedback electronically
Abstract
The invention discloses a digital hearing aid howling suppression system and a suppression method thereof, wherein the digital hearing aid howling suppression system comprises a preprocessor, a postprocessor, a subband modulator and a variable step length adaptive filter, wherein the preprocessor is used for processing a far-end signal of a hearing aid
And near-end signal
And output pre-processed
And
to a variable step size adaptive filter, wherein the far-end signal
The near-end signal is used as the signal before being output to the DAC of the hearing aid
For the signal output by the ADC of the hearing aid, the post-processor is arranged to process the estimated echo-free signal output by the variable-step-size adaptive filter
To restore the change of the preprocessor to the tone quality and sampling rate of the original signal; the invention realizes the adjustment of the frequency response of the signal and the emphasis of the effective signalThe function of the suppressing system and the suppressing method thereof can also improve the voice quality, can flexibly control the iteration speed of the system so as to obtain higher system robustness, have better using effect and are suitable for wide popularization and use.
Description
Technical Field
The invention relates to the technical field of digital hearing aids, in particular to a digital hearing aid howling suppression system and a suppression method thereof.
Background
With the continuous development of computer technology and audiology, digital hearing aids are becoming more and more popular. The digital hearing aid converts the acoustic signals into digital signals, and a built-in processor of the digital hearing aid can be used for processing and playing the acquired signals so as to improve the listening comfort and the definition.
Because the hearing aid generally needs to amplify sound, and the microphone and the receiver are close to each other, the sound amplified by the hearing aid is output from the receiver and then transmitted to the microphone through the space, and is circularly amplified to a saturated output state, so that howling is caused.
Currently, there are reports showing that about 24% of hearing aid wearers consider the feedback of the hearing aid to be a serious impact on the normal use of the hearing aid; howling suppression systems are therefore also an important component of digital hearing aid systems. The existing hearing aid howling suppression system generally has the following methods: one is automatic gain control, and the other is adaptive filtering echo cancellation; the automatic gain control method usually reduces the gain when howling occurs, but the method has effective hysteresis and also has the problem of wrong discrimination, thereby causing the situation of voice blockage; while the adaptive filtering echo cancellation method is generally used more commonly, the adaptive filtering echo cancellation method is easy to adapt slowly under the condition that far-end and near-end signals are strongly correlated in the using process of a hearing aid, and is easy to disperse so as to destroy the voice quality; therefore, it is necessary to design a digital hearing aid howling suppression system and a suppression method thereof.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a digital hearing aid howling suppression system and a suppression method thereof for better and effectively solving the problems of poor howling suppression capability, slow response speed and easy damage to voice quality in the existing method, has the effects of adjusting the frequency response of signals and emphasizing effective signals, and can flexibly control the iteration speed of the system so as to obtain higher system robustness.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a digital hearing aid howling suppression system comprises a preprocessor, a postprocessor, a subband modulator and a variable-step-size adaptive filter, wherein the preprocessor is used for processing a far-end signal u (n) and a near-end signal v (n) of a hearing aid and outputting the preprocessed far-end signal p _ u (n) and p _ v (n) to the variable-step-size adaptive filter, the far-end signal u (n) is a signal before being output to a DAC (digital-to-analog converter) of the hearing aid, the near-end signal v (n) is a signal output by an ADC (analog-to-digital converter) of the hearing aid, the preprocessor internally comprises an up-sampling filter and a pre-emphasis filter, the up-sampling filter is used for improving the sensitivity of the howling suppression system, and the pre-emphasis filter is used for reducing the intensity of low-frequency signals which do not generate howling and the intensity of high-frequency signals which are easy to generate the howling;
the post processor is used for processing the estimation output by the variable step size adaptive filter without the echo signal p _ e (n) so as to restore the change of the sound quality and the sampling rate of the original signal by the pre-processor, and outputting the obtained signal e (n) to the hearing aid algorithm module, and the post processor internally comprises a down sampling filter and a de-emphasis filter, wherein the down sampling filter is used for restoring the sampling rate of the signal so as to reduce the calculation amount of the whole hearing aid system, and the de-emphasis filter is used for restoring the signal frequency response changed by the pre-emphasis filter;
the subband modulator is used for modulating the signal x (n) output by the hearing aid algorithm module, outputting the signal to the DAC for playing and outputting the signal to the preprocessor as a far-end signal u (n), wherein the signal x (n) is output by the hearing aid algorithm module and is subjected to subband modulation operation before being output by the DAC, and the specific operation process is as follows,
s1, performing sub-band analysis on X (N) and obtaining N sub-bands X with different frequencies n (1),X n (2),…,X n (N), then, the frequency modulation is carried out on each sub-band as shown in the formula (1),
wherein k is 0,1,2 …, and N-1 is a sub-band sequence number;
is the phase modulation factor, e is the base of the natural logarithm, j is the imaginary unit,
to lose phase;
s2, the loss phase is composed of two parts, one part is constant offset f v (k) A defined time-invariant component, the other part being the modulation frequency f m And a periodic time-varying component of amplitude a (k), as shown in equation (2),
wherein, the first and the second end of the pipe are connected with each other,
is the initial phase;
s3, modulating each sub-band Z after frequency modulation n (k) Synthesizing to obtain a far-end signal u (n) and outputting the far-end signal u (n) to a DAC for playing;
the variable step length self-adapting filter is used for filtering the output signals p _ u (n) and p _ v (n) of the preprocessor, outputting the estimated echo-free signals p _ e (n) to the postprocessor, and then calculating the time-varying step length parameters and self-adaptively updating the filter weights.
Preferably, the subband modulator includes a subband analyzing module, a subband modulating module, and a subband synthesizing module, wherein the subband analyzing module is configured to divide an input signal into N subbands with different frequency components, the subband modulating module is configured to perform frequency modulation on a signal in each subband, and the subband synthesizing module is configured to synthesize and output each subband signal after frequency modulation.
Preferably, the step-size controller is included in the variable-step-size adaptive filter, and the step-size controller is configured to calculate an updated time-varying step size μ (n) of the adaptive filter according to the input signal to control the adaptation speed of the adaptive filter.
The method for suppressing the digital hearing aid howling suppression system comprises the following steps,
preprocessing a near-end signal v (n) and a far-end historical signal u (n) of the digital hearing aid, and outputting the processed signals p _ v (n) and p _ u (n) to a variable-step-size adaptive filter;
step (B), utilizing the step-variable self-adaptive filter to filter and output the preprocessed far-end signal p _ u (n) to estimate an echo signal
Step (C), counting the preprocessed far-end signal p _ u (n) and estimating the short-time energy of the signal p _ e (n) without echo, and carrying out self-adaptive updating on the filter weight according to the condition;
step (D), post-processing the estimated echo-free signal p _ e (n) and obtaining a signal e (n) required by a digital hearing aid algorithm;
and (E) performing subband modulation operation on the signals before the signals x (n) are output to the DAC by the hearing aid algorithm to finish suppression operation.
The method for suppressing digital hearing aid howling suppression system comprises the following steps of (A), preprocessing a near-end signal v (n) and a far-end historical signal u (n) of the digital hearing aid, and outputting the processed signals p _ v (n) and p _ u (n) to a variable step size adaptive filter,
step (A1), if the sampling rate of the near-end signal v (n) and the far-end historical signal u (n) is less than 20K, the original signal is up-sampled to more than 20K;
step (A2), pre-emphasis filter H is adopted pre (z) pre-emphasis is performed on the signal with filter characteristics that emphasize medium and high frequencies and attenuate low frequencies.
The method for suppressing the digital hearing aid howling suppression system includes the step (B) of estimating an echo signal by using a variable-step-size adaptive filter to filter and output the preprocessed far-end signal p _ u (n)
As shown in the formula (3), the,
wherein w (n) ═ w n (1),w n (2),…,w n (L)}，w n () adaptive filter weights, W H (n) is the conjugate transpose of w (n), u (n) ═ p _ u (n-D-L), p _ u (n-D-L +1), …, p _ u (n-D-1) } is the preprocessed far-end signal vector corresponding to the D sample time delays, L is the adaptive filter order; then subtracting the estimated echo signal from the preprocessed near-end signal to obtain the estimated echo-free signal p _ e (n) as shown in formula (4),
the method for suppressing the digital hearing aid howling suppression system comprises the step (C) of counting the preprocessed far-end signals p _ u (n) and estimating the short-time energy of the echo-free signals p _ e (n) and performing adaptive update of the filter weight according to the conditions, which comprises the following steps,
step (C1), calculating the short-time energy of the preprocessed far-end signal as shown in formula (5),
u_nrg(n)＝‖U(n)‖ 2 (5)
wherein | U (n) | 2 U (n) { p _ u (n-D-L), p _ u (n-D-L +1), …, p _ u (n-D-1) The remote signal vector after the pretreatment corresponding to the time delay of the D sampling points is obtained, and L is the order of the self-adaptive filter;
step (C2), calculating and estimating the short-time energy without echo signal as shown in formula (6),
e_nrg(n)＝‖E(n)‖ 2 (6)
wherein | E (n) | 2 Is the square of the euclidean norm of the vector e (n) { p _ e (n-L +1), p _ e (n-L +2), …, p _ e (n) };
step (C3), calculating the feedback such as shown in formula (7),
step (C4), controlling the iteration step of the weight of the adaptive filter according to the short-time energy u _ nrg (n) and the signal-to-noise ratio eur (n) of the preprocessed far-end signal, if u _ nrg (n)<Th1 or eur (n)>Th2, the step size μ (n) is equal to 0, otherwise
The method comprises the following specific steps of,
step (C41), the step size controller counts the preprocessed far-end input signal p _ u (n) short-time energy u _ nrg (n), if the short-time energy u _ nrg (n) of the far-end input signal is smaller than the acoustic feedback threshold Th1, the iteration step size μ (n) of the adaptive filter is equal to 0;
step (C42) of calculating and estimating the short-time energy e _ nrg (n) and the signal-to-noise ratio of the echo signal p _ e (n)
If eur (n) is greater than threshold Th2, then the adaptive filter iteration step size μ (n) is 0;
step (C43), other conditions use normalized adaptive step size
The adaptive filter weights are updated where ε is a constant greater than 0.
The invention has the beneficial effects that: the invention relates to a digital hearing aid howling suppression system, which adopts a preprocessor and a postprocessor to adjust the frequency response of signals and emphasize effective signals, so that the problem of low convergence speed of the current self-adaptive filtering method can be solved, then adopts a subband modulator to solve the problem that the filter of the current self-adaptive filtering method is easy to diverge so as to influence the voice tone quality, and adopts a variable-step self-adaptive method to flexibly control the iteration speed of the system so as to obtain higher system robustness.
Drawings
Fig. 1 is a block diagram of a digital hearing aid howling suppression system according to the present invention;
fig. 2 is a diagram comparing the performance of the howling suppression system of the digital hearing aid according to the present invention with that of the conventional method;
fig. 3 is a diagram comparing a voice signal processed by a suppression system of the present invention with a voice signal without a howling suppression system.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
As shown in fig. 1, a digital hearing aid howling suppression system of the present invention includes a preprocessor, a postprocessor, a subband modulator, and a variable step size adaptive filter; the preprocessor is used for processing a far-end signal u (n) and a near-end signal v (n) of the hearing aid and outputting the preprocessed p _ u (n) and p _ v (n) to a variable-step adaptive filter, wherein the far-end signal u (n) is a signal output to a DAC of the hearing aid, the near-end signal v (n) is a signal output by an ADC of the hearing aid, and the preprocessor internally comprises an up-sampling filter and a pre-emphasis filter, wherein the up-sampling filter is used for improving the sensitivity of a howling suppression system so as to quickly cope with the change of a feedback path; the pre-emphasis filter is used for reducing the strength of a low-frequency signal which does not generate howling and emphasizing the strength of a high-frequency signal which easily generates howling so as to improve the quality of a signal output to the variable-step-size adaptive filter;
and n is time, and the preprocessor performs resampling and filtering operation on the far-end signal and the near-end signal.
The post-processor is used for processing the estimation output by the variable step size adaptive filter without echo signal p _ e (n) to restore the change of the original signal tone quality and sampling rate of the pre-processor and outputting the signal e (n) to the hearing aid algorithm module, and the post-processor internally comprises a down-sampling filter and a de-emphasis filter, wherein the down-sampling filter is used for restoring the sampling rate of the signal to reduce the calculation amount of the whole hearing aid system, and the de-emphasis filter is used for restoring the signal frequency response changed by the pre-emphasis filter.
The subband modulator is used for modulating the signal x (n) output by the hearing aid algorithm module, outputting the signal to the DAC for playing and outputting the signal to the preprocessor as a far-end signal u (n), wherein the hearing aid algorithm module outputs the signal x (n) before outputting the signal to the DAC, the subband modulator is used for performing subband modulation operation on the signal, and the specific operation process is as follows,
s1, performing sub-band analysis on X (N) and obtaining N sub-bands X with different frequencies n (1),X n (2),…,X n (N), then, the frequency modulation is carried out on each sub-band as shown in the formula (1),
wherein k is 0,1,2 …, and N-1 is a sub-band sequence number;
is the phase modulation factor, e is the base of the natural logarithm, j is the imaginary unit,
to lose phase;
s2, the loss phase is composed of two parts, one part is constant offset f v (k) A defined time-invariant component, the other part being the modulation frequency f m And a periodic time-varying component of amplitude a (k), as shown in equation (2),
wherein the content of the first and second substances,
is the initial phase;
s3, modulating each sub-band Z after frequency modulation n (k) And synthesizing to obtain a far-end signal u (n) and outputting the far-end signal u (n) to the DAC for playing.
The variable step length self-adapting filter is used for filtering the output signals p _ u (n) and p _ v (n) of the pre-processor and outputting the estimated echo-free signals p _ e (n) to the post-processor, and then calculating the time-varying step length parameters and self-adaptively updating the filter weights.
Specifically, the subband modulator includes a subband analyzing module, a subband modulating module, and a subband synthesizing module, where the subband analyzing module is configured to divide an input signal into N subbands with different frequency components, the subband modulating module is configured to perform frequency modulation on a signal in each subband, and the subband synthesizing module is configured to synthesize and output each subband signal after frequency modulation.
Specifically, the step-size controller is internally included in the variable-step-size adaptive filter and is used for calculating the updating time-varying step size mu (n) of the adaptive filter according to the input signal so as to control the adaptation speed of the adaptive filter.
A method for suppressing a digital hearing aid howling suppression system comprises the following steps:
step (A), preprocessing a near-end signal v (n) and a far-end historical signal u (n) of the digital hearing aid, and outputting the processed signals p _ v (n) and p _ u (n) to a variable-step-size adaptive filter, wherein the specific steps are as follows,
step (A1), if the sampling rate of the near-end signal v (n) and the far-end historical signal u (n) is less than 20K, the original signal is up-sampled to more than 20K;
step (A2), using a pre-emphasis filter H pre (z) performing pre-emphasis operation on the signal, wherein the filter characteristic of the pre-emphasis operation is to strengthen medium and high frequencies and weaken low frequencies.
Step (B), utilizing a variable step length self-adaptive filter to carry out pretreatment on the far-end signal p _u (n) filtering the output of the estimated echo signal
As shown in the formula (3),
wherein w (n) { w n (1),w n (2),…,w n (L)}，w n (x) is the adaptive filter weight, W H (n) is the conjugate transpose of w (n), u (n) { p _ u (n-D-L), p _ u (n-D-L +1), …, p _ u (n-D-1) } is the preprocessed far-end signal vector corresponding to the D sample time delays, L is the adaptive filter order; then subtracting the estimated echo signal from the preprocessed near-end signal to obtain the estimated echo-free signal p _ e (n) as shown in formula (4),
step (C), counting the preprocessed far-end signal p _ u (n) and estimating the short-time energy of the echo-free signal p _ e (n) and carrying out the adaptive updating of the filter weight according to the conditions, which comprises the following specific steps,
step (C1), calculating the short-time energy of the preprocessed far-end signal as shown in formula (5),
u_nrg(n)＝‖U(n)‖ 2 (5)
wherein | U (n) | 2 The squared euclidean norm of the preprocessed far-end signal vector, u (n) ═ { p _ u (n-D-L), p _ u (n-D-L +1), …, p _ u (n-D-1) } is the preprocessed far-end signal vector corresponding to the D sample time delays, and L is the adaptive filter order;
step (C2), calculating and estimating the short-time energy without echo signal as shown in formula (6),
e_nrg(n)＝‖E(n)‖ 2 (6)
wherein | E (n) | 2 Is the square of the euclidean norm of the vector e (n) { p _ e (n-L +1), p _ e (n-L +2), …, p _ e (n) };
step (C3), calculating the feedback such as shown in formula (7),
step (C4), controlling the iteration step size of the weight of the adaptive filter according to the short-time energy u _ nrg (n) and the signal-to-return ratio eur (n) of the preprocessed far-end signal, if u _ nrg (n)<Th1 or eur (n)>Th2, the step size μ (n) is equal to 0, otherwise
The method comprises the following specific steps of,
step (C41), the step size controller counts the preprocessed far-end input signal p _ u (n) short-time energy u _ nrg (n), if the short-time energy u _ nrg (n) of the far-end input signal is smaller than the acoustic feedback threshold Th1, the iteration step size μ (n) of the adaptive filter is equal to 0;
step (C42) of calculating and estimating the short-time energy e _ nrg (n) and the signal-to-noise ratio of the echo signal p _ e (n)
If eur (n) is greater than the threshold Th2, the adaptive filter iteration step μ (n) is 0;
step (C43), other conditions use normalized adaptive step size
The adaptive filter weights are updated where ε is a constant greater than 0.
Step (D), post-processing the estimated echo-free signal p _ e (n) and obtaining a signal e (n) required by a digital hearing aid algorithm;
and (E) performing subband modulation operation on the signals before the signals x (n) are output to the DAC by the hearing aid algorithm to finish suppression operation.
To better illustrate the advantageous effects of the present invention, a specific embodiment of the present invention is described below. As shown in fig. 2, the same hearing aid adopts the conventional echo cancellation method and the adaptive filter weight unscheduled according to the technique of the present invention under the same environment, the unscheduled of the present invention is significantly lower and decreases faster, and has better howling suppression performance and response speed compared with the conventional method; as shown in fig. 3, which is a comparison graph of the output signal of the hearing aid without the howling suppression system and the output signal of the hearing aid adopting the technology of the present invention under the same condition, it can be seen that the hearing aid may rapidly generate howling at a certain time and cannot suppress the howling when the howling suppression system is absent, while when the technology of the present invention is adopted, the howling is not generated and the voice quality is not affected; the effectiveness of the inhibition system and inhibition method of the present invention is fully demonstrated.
In summary, in the digital hearing aid howling suppression system of the present invention, the near-end signal v (n) and the far-end historical signal u (n) of the digital hearing aid are preprocessed, and the preprocessed signals p _ v (n) and p _ u (n) are output to the step-size-variable adaptive filter; then, the preprocessed far-end signal p _ u (n) is filtered and output by utilizing a step-size-variable self-adaptive filter to estimate an echo signal
Then, counting the preprocessed far-end signal p _ u (n) and estimating the short-time energy of the signal p _ e (n) without echo, and carrying out self-adaptive updating on the filter weight according to the condition; then, carrying out post-processing on the estimated echo-free signal p _ e (n) to obtain a signal e (n) required by a digital hearing aid algorithm, and carrying out sub-band modulation operation on the signal before outputting a signal x (n) to a DAC (digital-to-analog converter) through the hearing aid algorithm to finish suppression operation; the invention adjusts the frequency response of the signal and emphasizes the effective signal by adopting the preprocessor and the postprocessor, thus improving the problem of low convergence speed of the current self-adaptive filtering method, then adopting the sub-band modulator to improve the problem of influence on voice tone quality caused by easy divergence of the filter of the current self-adaptive filtering method, and then adopting the variable-step self-adaptive method to flexibly control the iteration speed of the system so as to obtain higher system robustness.
The foregoing shows and describes the general principles, principal features and advantages of the 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 given by way of illustration of the principles of the present invention, but that various changes and modifications may be made without departing from the spirit and scope of the invention, and such changes and modifications are 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 digital hearing aid howling suppression system comprises a preprocessor, a postprocessor, a subband modulator and a variable-step-size adaptive filter, and is characterized in that: the preprocessor is used for processing a far-end signal u (n) and a near-end signal v (n) of the hearing aid and outputting preprocessed p _ u (n) and p _ v (n) to a step-variable adaptive filter, wherein the far-end signal u (n) is a signal output to a DAC of the hearing aid, the near-end signal v (n) is a signal output by an ADC of the hearing aid, the preprocessor internally comprises an upsampling filter and a pre-emphasis filter, the upsampling filter is used for improving the sensitivity of a howling suppression system, and the pre-emphasis filter is used for reducing the strength of low-frequency signals which do not generate howling and emphasizing the strength of high-frequency signals which easily generate the howling;
the post processor is used for processing the estimation output by the variable step size adaptive filter without the echo signal p _ e (n) so as to restore the change of the sound quality and the sampling rate of the original signal by the pre-processor, and outputting the obtained signal e (n) to the hearing aid algorithm module, and the post processor internally comprises a down sampling filter and a de-emphasis filter, wherein the down sampling filter is used for restoring the sampling rate of the signal so as to reduce the calculation amount of the whole hearing aid system, and the de-emphasis filter is used for restoring the signal frequency response changed by the pre-emphasis filter;
the subband modulator is used for modulating the signal x (n) output by the hearing aid algorithm module, outputting the signal to the DAC for playing and outputting the signal to the preprocessor as a far-end signal u (n), wherein the signal x (n) is output by the hearing aid algorithm module and is subjected to subband modulation operation before being output by the DAC, and the specific operation process is as follows,
s1, performing sub-band analysis on X (N) and obtaining N sub-bands X with different frequencies n (1),X n (2),…,X n (N), then, each sub-band is frequency modulated as shown in formula (1),
wherein k is 0,1,2 …, and N-1 is a sub-band number;
is the phase modulation factor, e is the base of the natural logarithm, j is the imaginary unit,
to lose phase;
s2, the loss phase is composed of two parts, one part is constant offset f v (k) A defined time-invariant component, the other part being the modulation frequency f m And a periodic time-varying component of amplitude a (k), as shown in equation (2),
wherein, the first and the second end of the pipe are connected with each other,
is the initial phase;
s3, modulating each sub-band Z after frequency modulation n (k) Synthesizing to obtain a remote signal u (n) and outputting the remote signal u (n) to a DAC for playing;
the variable step length self-adapting filter is used for filtering the output signals p _ u (n) and p _ v (n) of the pre-processor and outputting the estimated echo-free signals p _ e (n) to the post-processor, and then calculating the time-varying step length parameters and self-adaptively updating the filter weights.
2. A digital hearing aid howling suppression system according to claim 1, wherein: the subband modulator comprises a subband analyzing module, a subband modulating module and a subband synthesizing module, wherein the subband analyzing module is used for dividing an input signal into N subbands with different frequency components, the subband modulating module is used for carrying out frequency modulation on the signal in each subband, and the subband synthesizing module is used for synthesizing and outputting each subband signal after the frequency modulation.
3. A digital hearing aid howling suppression system according to claim 1, wherein: the step-size controller is internally included in the variable-step-size adaptive filter and used for calculating the updating time-varying step size mu (n) of the adaptive filter according to the input signal so as to control the adaptive speed of the adaptive filter.
4. A method for suppressing howling in a digital hearing aid according to any one of claims 1-3, wherein: comprises the following steps of (a) preparing a solution,
step (A), preprocessing a near-end signal v (n) and a far-end historical signal u (n) of a digital hearing aid, and outputting the processed signals p _ v (n) and p _ u (n) to a variable step size adaptive filter;
step (B), utilizing the step-variable self-adaptive filter to filter and output the preprocessed far-end signal p _ u (n) to estimate an echo signal
Step (C), counting the preprocessed far-end signal p _ u (n) and estimating the short-time energy of the echo-free signal p _ e (n), and carrying out self-adaptive updating on the filter weight according to the condition;
step (D), post-processing the estimated echo-free signal p _ e (n) and obtaining a signal e (n) required by a digital hearing aid algorithm;
and (E) before the hearing aid algorithm outputs the signal x (n) to the DAC, performing subband modulation operation on the signal to finish suppression operation.
5. The method for suppressing howling suppression in a digital hearing aid according to claim 4, wherein: step (A), preprocessing a near-end signal v (n) and a far-end historical signal u (n) of the digital hearing aid, and outputting the processed signals p _ v (n) and p _ u (n) to a variable-step-size adaptive filter, wherein the specific steps are as follows,
step (A1), if the sampling rate of the near-end signal v (n) and the far-end historical signal u (n) is less than 20K, the original signal is up-sampled to more than 20K;
step (A2), pre-emphasis filter H is adopted pre (z) performing pre-emphasis operation on the signal, wherein the filter characteristic of the pre-emphasis operation is to strengthen medium and high frequencies and weaken low frequencies.
6. The method for suppressing the howling suppression system of the digital hearing aid according to claim 4, wherein: step (B), using the variable step length adaptive filter to filter the preprocessed far-end signal p _ u (n) and output the estimated echo signal
As shown in the formula (3),
wherein w (n) { w n (1),w n (2),…,w n (L)}，w n (x) is the adaptive filter weight, W H (n) is the conjugate transpose of w (n), u (n) ═ p _ u (n-D-L), p _ u (n-D-L +1), …, p _ u (n-D-1) } is the preprocessed far-end signal vector corresponding to the D sample time delays, L is the adaptive filter order; then subtracting the estimated echo signal from the preprocessed near-end signal to obtain the estimated echo-free signal p _ e (n) as shown in the formula (4),
7. the method for suppressing the howling suppression system of the digital hearing aid according to claim 4, wherein: step (C), counting the preprocessed far-end signal p _ u (n) and estimating the short-time energy of the echo-free signal p _ e (n) and carrying out the adaptive updating of the filter weight according to the conditions,
step (C1), calculating the short-time energy of the preprocessed far-end signal as shown in formula (5),
u_nrg(n)＝‖U(n)‖ 2 (5)
wherein | U (n) | 2 The squared euclidean norm of the preprocessed far-end signal vector, u (n) ═ { p _ u (n-D-L), p _ u (n-D-L +1), …, p _ u (n-D-1) } is the preprocessed far-end signal vector corresponding to the D sample time delays, and L is the adaptive filter order;
step (C2), calculating and estimating the short-time energy without echo signal as shown in formula (6),
e_nrg(n)＝‖E(n)‖ 2 (6)
wherein | E (n) | 2 Is the square of the euclidean norm of vector e (n) { p _ e (n-L +1), p _ e (n-L +2), …, p _ e (n) };
step (C3), calculating the signal such as shown in formula (7),
step (C4), controlling the iteration step size of the adaptive filter weight according to the preprocessed far-end signal short-time energy u _ nrg (n) and the signal-back ratio eur (n), if u _ nrg (n) < Th1 or eur (n) > Th2, the step size mu (n) ═ 0, otherwise
The method comprises the following specific steps of,
step (C41), the step size controller counts the preprocessed far-end input signal p _ u (n) short-time energy u _ nrg (n), if the short-time energy u _ nrg (n) of the far-end input signal is smaller than the acoustic feedback threshold Th1, the iteration step size μ (n) of the adaptive filter is equal to 0;
step (C42) of calculating and estimating the short-time energy e _ nrg (n) and the signal-to-noise ratio of the echo signal p _ e (n)
If eur (n) is greater than threshold Th2, then the adaptive filter iteration step size μ (n) is 0;
step (C43), other conditions use normalized adaptive step size
The adaptive filter weights are updated where ε is a constant greater than 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210463650.4A CN114584909B (en) | 2022-04-29 | 2022-04-29 | Digital hearing aid howling suppression system and suppression method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210463650.4A CN114584909B (en) | 2022-04-29 | 2022-04-29 | Digital hearing aid howling suppression system and suppression method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114584909A CN114584909A (en) | 2022-06-03 |
| CN114584909B true CN114584909B (en) | 2022-07-26 |
Family
ID=81778530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210463650.4A Active CN114584909B (en) | 2022-04-29 | 2022-04-29 | Digital hearing aid howling suppression system and suppression method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114584909B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006262098A (en) * | 2005-03-17 | 2006-09-28 | Yamaha Corp | Howling canceller |
| JP2012165281A (en) * | 2011-02-08 | 2012-08-30 | Nippon Telegr & Teleph Corp <Ntt> | Echo cancelling method, echo cancelling device and program |
| CN106782592A (en) * | 2016-12-27 | 2017-05-31 | 中山大学花都产业科技研究院 | A kind of echo and the system and method uttered long and high-pitched sounds for eliminating network sound transmission |
| CN109215675A (en) * | 2017-07-05 | 2019-01-15 | 上海谦问万答吧云计算科技有限公司 | A kind of method, device and equipment of chauvent's criterion |
| CN111916099A (en) * | 2020-10-13 | 2020-11-10 | 南京天悦电子科技有限公司 | Adaptive echo cancellation device and method for variable-step hearing aid |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5091952A (en) * | 1988-11-10 | 1992-02-25 | Wisconsin Alumni Research Foundation | Feedback suppression in digital signal processing hearing aids |
| CA2481629A1 (en) * | 2004-09-15 | 2006-03-15 | Dspfactory Ltd. | Method and system for active noise cancellation |
| KR100827146B1 (en) * | 2004-09-20 | 2008-05-02 | 삼성전자주식회사 | Method and apparatus for elimination acoustic echo in mobile terminal |
| US9578426B2 (en) * | 2015-03-20 | 2017-02-21 | Sonova Ag | Method for feedback cancelling in hearing devices and hearing device with a feedback canceller |
| CN110611871B (en) * | 2019-09-09 | 2020-12-01 | 惠州市锦好医疗科技股份有限公司 | Howling suppression method and system for digital hearing aid and special DSP |
| CN112954576B (en) * | 2021-02-03 | 2022-03-04 | 天津大学 | Digital hearing aid howling detection and suppression algorithm based on filter bank and hardware implementation method |
-
2022
- 2022-04-29 CN CN202210463650.4A patent/CN114584909B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006262098A (en) * | 2005-03-17 | 2006-09-28 | Yamaha Corp | Howling canceller |
| JP2012165281A (en) * | 2011-02-08 | 2012-08-30 | Nippon Telegr & Teleph Corp <Ntt> | Echo cancelling method, echo cancelling device and program |
| CN106782592A (en) * | 2016-12-27 | 2017-05-31 | 中山大学花都产业科技研究院 | A kind of echo and the system and method uttered long and high-pitched sounds for eliminating network sound transmission |
| CN109215675A (en) * | 2017-07-05 | 2019-01-15 | 上海谦问万答吧云计算科技有限公司 | A kind of method, device and equipment of chauvent's criterion |
| CN111916099A (en) * | 2020-10-13 | 2020-11-10 | 南京天悦电子科技有限公司 | Adaptive echo cancellation device and method for variable-step hearing aid |
Non-Patent Citations (4)
| Title |
|---|
| An Improved Practical State-Space FDAF With Fast Recovery of Abrupt Echo-Path Changes;Tao Jiang;《IEEE Access》;20190412;全文 * |
| Piecewise-Linear Frequency Shifting Algorithm for Frequency Resolution Enhancement in Digital Hearing Aids;Qingyun Wang;《applied sciences》;20170329;全文 * |
| 应用于助听器反馈抑制系统的变步长归一化子带自适应滤波算法;王侠;《东南大学学报》;20150531;全文 * |
| 语音系统中的子带自适应回声消除技术;蔡宇等;《仪器仪表学报》;20130715(第07期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114584909A (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8000482B2 (en) | Microphone array processing system for noisy multipath environments | |
| EP1529282B1 (en) | Method and system for processing subband signals using adaptive filters | |
| EP1208689B1 (en) | Acoustical echo cancellation device | |
| US7773759B2 (en) | Dual microphone noise reduction for headset application | |
| JP5049277B2 (en) | Method and system for clear signal acquisition | |
| EP1936939B1 (en) | Low complexity echo compensation | |
| US9800734B2 (en) | Echo cancellation | |
| EP2186312B1 (en) | Acoustic echo cancellation based on noise environment | |
| US9992572B2 (en) | Dereverberation system for use in a signal processing apparatus | |
| CN1223166C (en) | Methods and apparatus for improved sub-band adaptive filtering in echo cancellation systems | |
| US8320575B2 (en) | Efficient audio signal processing in the sub-band regime | |
| US20080031467A1 (en) | Echo reduction system | |
| JP2002541753A (en) | Signal Noise Reduction by Time Domain Spectral Subtraction Using Fixed Filter | |
| US20180308503A1 (en) | Real-time single-channel speech enhancement in noisy and time-varying environments | |
| US8306821B2 (en) | Sub-band periodic signal enhancement system | |
| JPH09307625A (en) | Sub band acoustic noise suppression method, circuit and device | |
| Bulling et al. | Stepsize Control for Acoustic Feedback Cancellation Based on the Detection of Reverberant Signal Periods and the Estimated System Distance. | |
| Yang | Multilayer adaptation based complex echo cancellation and voice enhancement | |
| CN114584909B (en) | Digital hearing aid howling suppression system and suppression method thereof | |
| JP2002541529A (en) | Reduction of signal noise by time domain spectral subtraction | |
| EP1305975B1 (en) | Adaptive microphone array system with preserving binaural cues | |
| CA2437477C (en) | Method and system for processing subband signals using adaptive filters | |
| CN111836165A (en) | Compensation method for frequency response curve of electroacoustic device in active noise reduction system | |
| CN113347536B (en) | Acoustic feedback suppression algorithm based on linear prediction and sub-band adaptive filtering | |
| Bekrani et al. | Neural network based adaptive echo cancellation for stereophonic teleconferencing application |
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 |