CN102740214A - Howling suppression method based on feedback signal spectrum estimation - Google Patents

Abstract

The invention relates to a howling suppression method based on feedback signal spectrum estimation. The method comprises the following steps: firstly, carrying out framing, windowing and Fourier transform on a signal collected by a microphone and a loudspeaker output signal respectively so as to obtain a corresponding frequency domain signal; then, according to the microphone frequency domain signal and the loudspeaker output frequency domain signal, calculating a cross-power spectrum between the microphone and the loudspeaker output signal and a power spectrum of the loudspeaker output signal; dividing the cross-power spectrum by the power spectrum so as to obtain estimation of a feedback path transfer function amplitude spectrum; using the obtained estimation of the feedback path transfer function spectrum amplitude to calculate a current frame feedback signal amplitude spectrum; and then using the microphone signal amplitude spectrum and the obtained feedback signal amplitude spectrum to calculate a gain function; finally, according to the obtained gain function, carrying out correction on the microphone frequency domain signal so as to realize howling suppression. By using the method of the invention, convergence and tracking speeds are fast, robustness is good and algorithm complexity is low, which is good for real-time realization.

Description

A kind of inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation

Technical field

The inhibition field that the present invention relates to utter long and high-pitched sounds, particularly a kind of inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation.

Background technology

The document that is suppressed at of uttering long and high-pitched sounds also is called as " acoustic feedback inhibition ", " acoustic feedback control ", " feedback is eliminated " etc., and what they were expressed is with a kind of meaning.As shown in Figure 1, Fig. 1 is the generation schematic diagram of uttering long and high-pitched sounds.S (n) representes near-end speech, and this is to need amplifying signal; D (n) expression feeds back to microphone from the sound that loud speaker sends through feedback path 102 once more, the feedback signal of being picked up by microphone.D (n) is not the signal that expectation receives, and need curb it.Forward direction processing module 104 is used for accomplishing the functions such as inhibition and automatic gain control of uttering long and high-pitched sounds.Input signal s (n) is G (z)/(1-G (z) F (z)) to the transfer function between the output signal y (n), if satisfy (a) simultaneously | G (z) F (z) | >=1; (b) ∠ G (z) F (z)=2 π n, n ∈ N; Then this system will become unstable, on the Frequency point that satisfies condition, will utter long and high-pitched sounds.Using one of public address system purpose is exactly in order to improve system gain, and the wearer of hearing aids also hopes and can be amplified to desired level to sound, the amount of gain of sound reinforcement system of having uttered long and high-pitched sounds down phenomenon limits.Therefore, must take certain measure elimination feedback to avoid the generation of uttering long and high-pitched sounds.In the public sound reinforcement system because teller's position may change frequently.For example: teacher gives lessons in the public address system of usefulness in the classroom; Microphone is placed on teacher's the clothes and the position of teacher's health when giving lessons is not fixed; This just causes the continuous variation of feedback path; Corresponding solution must be considered this point and can follow the tracks of this variation fast, otherwise will occur uttering long and high-pitched sounds.

The inhibition of uttering long and high-pitched sounds is the research topic of a hot topic.Present solution mainly contains:

(1) shift frequency method.The shift frequency method reaches the purpose of uttering long and high-pitched sounds and suppressing through the destruction phase condition that produces of uttering long and high-pitched sounds.Yet uttering long and high-pitched sounds of this method suppresses effect and bad, usually controlled uttering long and high-pitched sounds of a Frequency point, and system can produce at other Frequency point again and utter long and high-pitched sounds.And the shift frequency method has noticeable influence to voice quality, and research and real system test shows use shift frequency method can obtain the gain of 6dB at most.

(2) trapper method.The trapper method is at first sought the spectrum position that feedback possibly take place or take place, and these strong excessively frequency contents in the signal is attenuated to reach the feedback inhibition effect then.But in real system, might there be a plurality of feedback frequency points.Therefore, trapper method effect is limited.The use trapper method of having reported can provide the gain of 3dB-8dB.

(3) sef-adapting filter method.The operation principle of sef-adapting filter method is again from the come in transmission characteristic of a such propagation ducts of microphone after utilizing filter to come simulated sound to send through loud speaker; Make that the signal that comes out from filter is consistent with the feedback signal that microphone collects, from the input signal of microphone, cut this part signal then and reach the purpose that elimination is fed back.As shown in Figure 2, Fig. 2 is the inhibition system block diagram of uttering long and high-pitched sounds based on adaptive filter algorithm.The parameter of adaptive-filtering unit 206 must be adjusted to the actual transmission paths characteristic approaching, and the just more approaching actual feedback signal of the signal that its simulation obtains just can obtain the inhibition effect of better uttering long and high-pitched sounds.This method is the same with the method principle that echo cancelltion is adopted; But different with echo cancelltion is under the feedback occasion; System always is in the state that is called as " both-end intercommunication "; Thereby influenced the convergence rate of sef-adapting filter, and the filter coefficient that estimates has partially, this method is difficult to be applicable to that feedback path is always in situation about constantly changing.

In addition, the sef-adapting filter method is divided into discontinuous the utter long and high-pitched sounds inhibition and the inhibition of uttering long and high-pitched sounds continuously.The discontinuous major defect that suppresses of uttering long and high-pitched sounds is that this method need be interrupted normal input speech signal, thereby the intelligibility of voice is affected, thereby is can not be received in a lot of occasions.Suppress to use continuous input signal to come estimated feedback signal and utter long and high-pitched sounds continuously, but convergence rate is very slow.

In sum, the present inhibition method of uttering long and high-pitched sounds robustness is bad, when feedback path changes, all occurs of short duration uttering long and high-pitched sounds probably.

Summary of the invention

The objective of the invention is to,, propose a kind of inhibition method of uttering long and high-pitched sounds, thereby it is good to reach robustness, convergence and the fast purpose of tracking velocity based on the feedback signal spectrum estimation in order to address the above problem.

For realizing the foregoing invention purpose, the present invention proposes a kind of inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation, and these method concrete steps comprise:

Step 1): the signal x (n) that respectively microphone is gathered and speaker output signal y (n) carry out branch frame, windowing and Fourier transform obtain the microphone frequency domain signal X (i, k) with loud speaker export frequency-region signal Y (i, k);

Step 2): (i, k) (i k) calculates crosspower spectrum φ between microphone and the speaker output signal to the microphone frequency domain signal X that obtains according to described step 1) with loud speaker output frequency-region signal Y _Yx(i is k) with the power spectrum φ of speaker output signal _Yy(i, k); And with crosspower spectrum φ _Yx(i is k) divided by power spectrum φ _Yy(i k) obtains the estimation of feedback path transfer function amplitude spectrum

Step 3): exploiting the said step 2) obtained in feedback path of transfer function of amplitude spectrum from the estimated

calculate get the current frame feedback signal amplitude spectrum of

Step 4): utilize the microphone signal amplitude spectrum | X (i; K) | and feedback signal amplitude spectrum

the calculating acquisition gain function G of described step 3) acquisition (i, k);

Step 5): (i k) comes that (i k) revises and realizes uttering long and high-pitched sounds inhibition to the microphone frequency domain signal X to the gain function G that obtains according to described step 4).

This method also further comprises:

Step 6): the said step 5) to obtain a corrected frequency domain signal after processing the microphone to obtain a final output signal

and the output signal

de-correlation processing; and

Step 7): described step 6) process result is carried out gain-adjusted.

Described step 2) concrete steps of acquisition feedback path transfer function amplitude spectrum comprise:

21): adopt the level and smooth method of single order autoregression, obtain crosspower spectrum φ according to formula (1) _Yx(i, k);

φ _yx(i，k)＝α _Cφ _yx(i-1，k)+(1-α _C)|Y ^*(i，k)X(i，k)| (1)

Wherein, α _CBe smoothing factor, 0≤α _C＜1;

22): adopt the level and smooth method of single order autoregression, obtain power spectrum φ according to formula (2) _Yy(i, k);

φ _yy(i，k)＝α _Rφ _yy(i-1，k)+(1-α _R)Y ^*(i，k)Y(i，k) (2)

Wherein, α _RBe smoothing factor, 0≤α _R＜1;

23): utilize crosspower spectrum φ _Yx(i is k) with power spectrum φ _Yy(i k) obtains feedback path transfer function amplitude spectrum according to formula (3)

$\left|{\hat{F}}^{\′}(i,k)\right|=\frac{{\mathrm{\φ}}_{\mathrm{yx}}(i,k)}{{\mathrm{\φ}}_{\mathrm{yy}}(i,k)+\mathrm{\δ}}---\left(3\right)$

Wherein, δ is that to avoid the denominator of formula (3) be 0 to a very little positive number.

Described step 3) is calculated according to formula (4) and is obtained present frame feedback signal amplitude spectrum

$\left|\hat{X}(i,k)\right|=\mathrm{\β}\left|{\hat{F}}^{\′}(i,k)\right|\left|X(i,k)\right|---\left(4\right)$

Wherein, β is an adjustable parameter, carries out this parameter of choose reasonable according to actual conditions.

The method of described step 4) calculated gains function comprises: based on priori signal feedback ratio method with based on check back signal feedback ratio method.

Described step 4) based on check back signal feedback ratio method according to formula (5) calculate to obtain gain function G (i, k);

$G(i,k)={\left[\frac{\max ({\left|X(i,k)\right|}^{\mathrm{\α}}-{\left|\hat{X}(i,k)\right|}^{\mathrm{\α}},0)}{{\left|X(i,k)\right|}^{\mathrm{\α}}}\right]}^{\frac{1}{\mathrm{\α}}}---\left(5\right)$

Wherein, α＞0.For reducing computation complexity, α gets 1 or 2.

(i k) revises and realizes uttering long and high-pitched sounds inhibition described step 5) to the microphone frequency domain signal X according to formula (6);

$\hat{S}(i,k)=X(i,k)G(i,k)---\left(6\right).$

Described step 7) adopts the gain of regulating digital signal or the gain of regulating the analogue system power amplifier.The invention has the advantages that compared with prior art, the present invention has good robustness; Initial convergence speed is fast and when feedback path changes, can follow the tracks of this variation rapidly and can not utter long and high-pitched sounds; Because all operations of the present invention all is that frequency domain carries out, therefore has very low computation complexity.

Description of drawings

Fig. 1 is the generation schematic diagram of uttering long and high-pitched sounds;

Fig. 2 is traditional inhibition system block diagram of uttering long and high-pitched sounds based on adaptive filter algorithm;

Fig. 3 is a feedback path estimating system block diagram of the present invention;

Fig. 4 is the inhibition method of the uttering long and high-pitched sounds structured flowchart based on the feedback signal spectrum estimation of the present invention;

Fig. 5 is the block diagram of feedback spectrum estimation module 410 of the present invention;

Fig. 6 is a feedback path impulse response figure of actual measurement;

Fig. 7 is the amplitude spectrum of actual feedback path transfer function and the amplitude spectrum comparison diagram of the feedback path transfer function that the present invention estimates out.

Embodiment

Below in conjunction with accompanying drawing and specific embodiment the present invention is carried out detailed explanation.

It is good to the objective of the invention is to have proposed a kind of robustness, and convergence and tracking velocity are fast, when effectively inhibition is uttered long and high-pitched sounds, guarantees the inhibition method of uttering long and high-pitched sounds that voice quality is not suffered a loss.

In order to achieve the above object, the technical scheme taked of the present invention is following:

A kind of inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation, concrete steps comprise:

Step 1): respectively microphone signal and the speaker output signal that collects carried out windowing process and Fourier transform, time-domain signal is transformed to frequency domain;

Step 2): utilize the crosspower spectrum of level and smooth microphone signal and speaker output signal to obtain the amplitude spectrum of feedback path, and utilize the amplitude spectrum of feedback path to calculate the amplitude spectrum of present frame feedback signal divided by the power spectrum of level and smooth speaker output signal;

Step 3): utilize microphone signal amplitude spectrum and feedback signal amplitude spectrum calculated gains function;

Step 4): the microphone signal frequency spectrum is revised;

Step 5): revised microphone signal frequency spectrum is carried out inverse Fourier transform obtain time domain output;

Step 6): optional) output to step 5) applies a de-correlation block;

Step 7): optional) output to step 6) applies a gain.

In technique scheme, further, the pattern that adopts frame to handle described in the step 1) is carried out signal processing, all handle and program all to be based on frame be unit.Need carry out windowing process to the frame microphone signal that receives and the output signal of loud speaker, use FFT to transform to frequency domain then.

In technique scheme, further, step 2) at first needs the amplitude spectrum of estimation feedback path transfer function described in, and utilize its to come further to calculate amplitude spectrum of present frame feedback signal.The crosspower spectrum absolute value that the amplitude spectrum of feedback path transfer function utilizes level and smooth microphone signal and speaker output signal obtains divided by the power spectrum of level and smooth speaker output signal.

In above-mentioned technical scheme, further, the said gain function of step 3) is based on that the check back signal feedback ratio calculates.

In above-mentioned technical scheme, further, step 4) only is meant the frequency spectrum correction of microphone its amplitude spectrum is revised, and keeps its phase spectrum constant.

In above-mentioned technical scheme, further, step 5) is revised microphone frequency spectrum to be carried out inverse Fourier transform obtain time-domain signal, and the output signal with former frame carries out overlapping stack as the output signal of uttering long and high-pitched sounds and suppressing then.

In above-mentioned technical scheme, further, the described de-correlation block of step 6) is with the correlation that removes between the near end signal that speaker output signal and microphone receive.

In above-mentioned technical scheme, further, the described gain of step 7) both can be added in the power amplifier end that also can be added in analogue system on the digital signal of digital system.

As shown in Figure 4, Fig. 4 is the inhibition method of the uttering long and high-pitched sounds structured flowchart based on the feedback signal spectrum estimation of the present invention.Wherein, s (n) representes near end signal, and d (n) representes feedback signal, the signal that x (n) expression microphone collects, the signal of y (n) expression loud speaker output.

Step 1): respectively microphone signal and the speaker output signal that collects carried out windowing process and Fourier transform, time-domain signal is transformed to frequency domain.

The microphone signal x (n) that collects is carried out behind the serial to parallel conversion through after windowing module 402 and the Fourier transform module 404 time-domain signal being transformed to frequency domain; Speaker output signal y (n) is carried out behind the serial to parallel conversion through after windowing module 402 and the Fourier transform module 404 time-domain signal being transformed to frequency domain.The i frame discrete Fourier transform of y (n) be designated as Y (i, k), the i frame discrete Fourier transform of x (n) be designated as X (i, k).

Step 2): the crosspower spectrum that utilizes level and smooth microphone signal x (n) and speaker output signal y (n) obtains the amplitude spectrum of feedback path divided by the power spectrum of level and smooth speaker output signal y (n), and utilizes the amplitude spectrum of feedback path to calculate the amplitude spectrum of present frame feedback signal.

As shown in Figure 3, Fig. 3 is a feedback path estimating system block diagram of the present invention.The signal that microphone picks up is x (n)=d (n)+s (n), and d (n) is that speaker output signal y (n) obtains through feedback path transfer function F (z) filtering.We are objective definition function E{e ²(n) }=E{|x (n)-y (n) * f (n) | ², * representes convolution here, the time-domain pulse response of f (n) expression F (z).Minimizing Wei Na that this target function can obtain time domain, to separate form following:

$\hat{f}\left(n\right)=R_{\mathrm{yy}}^{-1}{r}_{\mathrm{yx}}---\left(1\right)$

Wherein, R _YyThe autocorrelation matrix of expression y (n), r _YxThe cross-correlation matrix of expression y (n) and x (n).(1) is transformed to frequency domain to be obtained

$F^{\′}(i,k)=\frac{P_{\mathrm{yx}}(i,k)}{P_{\mathrm{yy}}(i,k)}---\left(2\right)$

Here P _Yx(i, the k) crosspower spectrum of expression x (n) and y (n), P _Yy(i, k) power spectrum of expression y (n).If the delay long enough between input signal s (n) and output signal y (n) just can well carry out decorrelation to the such stationary signal in short-term of voice, can think that promptly correlation between s (n) and the y (n) has become very a little less than, P so _Ys(i, k) ≈ 0, so P _Yx(i, k)=P _Ys(i, k)+P _Yd(i, k) ≈ P _Yd(i, k), thereby (2) formula of utilization can effectivelyly estimate the frequency spectrum of feedback path transfer function.The present invention is actual utilize be F ' (i, amplitude spectrum k) | F ' (i, k) |.As shown in Figure 5, Fig. 5 is the block diagram of feedback spectrum estimation module 410 of the present invention.

Crosspower spectrum smooth unit 502 is calculated the crosspower spectrum absolute value of x (n) and y (n), and carries out smoothly,

φ _yx(i，k)＝α _Cφ _yx(i-1，k)+(1-α _C)|Y ^*(i，k)X(i，k)| (3)

Here use * to represent complex conjugate.We have adopted signal period figure have been carried out autoregression smoothly as the estimation of power spectrum or crosspower spectrum, and this mainly is because this method is simple and easy the realization, yet this patent is not limited to this method, and the method for other any estimated power spectrums can be used.

Power spectrum smooth unit 504 is calculated the power spectrum of y (n) and is carried out smoothly,

φ _yy(i，k)＝α _Rφ _yy(i-1，k)+(1-α _R)Y ^*(i，k)Y(i，k) (4)

Here α _CAnd α _RBe smoothing factor, its value satisfies 0≤α _C＜1 and 0≤α _R＜1.

Feedback transfer function estimation unit 506 calculates feedback path transfer function amplitude spectrum, is expressed as:

$\left|{\hat{F}}^{\′}(i,k)\right|=\frac{{\mathrm{\φ}}_{\mathrm{yx}}(i,k)}{{\mathrm{\φ}}_{\mathrm{yy}}(i,k)+\mathrm{\δ}}---\left(5\right)$

δ is a positive number that absolute value is very little in the formula (5), and being used for the divisor of the formula that prevents (5) is 0.

As shown in Figure 6, Fig. 6 is the feedback path impulse response figure of an actual measurement.As shown in Figure 7, Fig. 7 is the amplitude spectrum of actual feedback path transfer function and the amplitude spectrum comparison diagram of the feedback path transfer function that the present invention estimates out.The amplitude spectrum that can find out the feedback path transfer function that the present invention estimates can well approach the amplitude spectrum of real feedback path transfer function.

Present frame feedback signal amplitude spectrum computing unit 508 is used for estimating the spectrum component of present frame feedback signal, is expressed as

$\left|\hat{X}(i,k)\right|=\mathrm{\β}\left|{\hat{F}}^{\′}(i,k)\right|\left|Y(i,k)\right|---\left(6\right)$

In the formula (6), β is a parameter that the user is adjustable, if the amplitude spectrum of feedback signal is crossed and to have been estimated then need make β＜1, otherwise need be made β＞1 if the amplitude spectrum of feedback signal owes to have estimated then.Select big compensating factor just can obtain bigger amount of gain but relative meeting the near-end speech quality is had certain influence; The voice quality of selecting little compensating factor to obtain still may reduce the amount of gain of sound reinforcement system, should reasonably accept or reject according to actual conditions.

In sum, we have just estimated the amplitude spectrum of feedback signal through type (6), have accomplished the function of feedback signal spectrum estimation module 410.

Step 3): utilize microphone signal amplitude spectrum and feedback signal amplitude spectrum calculated gains function;

The gain function of gain calculation module 412 can utilize spectrum-subtraction to obtain, and its expression formula is:

$G(i,k)={\left[\frac{\max ({\left|X(i,k)\right|}^{\mathrm{\α}}-{\left|\hat{X}(i,k)\right|}^{\mathrm{\α}},0)}{{\left|X(i,k)\right|}^{\mathrm{\α}}}\right]}^{\frac{1}{\mathrm{\α}}}---\left(7\right)$

α representes a factor of spectrum-subtraction in the following formula, and α=1 is based on the amplitude spectrum subtraction, α=2nd, and the power spectrum subtraction is the Wiener filtering method, α also can select other numerical value, α＞0.Generally select α=1 or α=2 in order to reduce computation complexity in the reality.

Though the selection of gain function also has other form, as adopted in the document [1] gain function based on the priori signal to noise ratio ([1] Y.Ephraim and D.Malah, " Speech enhancement using a minimum mean squareerror short-time spectral amplitude estimator; " IEEE Trans.on Acoust.; Speech, SignalProcessing, vol.ASSP-32; Pp.1109-1121; Dec.1984), find to use posterior signal feedback ratio can obtain better result through testing us, thereby the present invention recommends use formula (7) to come the calculated gains function for the inhibition system that utters long and high-pitched sounds.

Step 4) is revised the microphone signal frequency spectrum.

The input/output relation of frequency spectrum correcting module 414 is:

$\hat{S}(i,k)=X(i,k)G(i,k)---\left(8\right)$

The microphone signal frequency spectrum after the feedback signal has been eliminated in

expression.

Step 5) is carried out inverse Fourier transform to the microphone signal frequency spectrum and is obtained output.

416 pairs of present frames of inverse Fourier transform module

carry out inverse Fourier transform and obtain present frame output, carry out overlapping stack with former frame output then and obtain final output signal

Step 6) is optional) output of step 5) is applied a de-correlation block 418.

Like step 2) delay between said input signal s (n) and output signal y (n) is 3 times frame length, promptly corresponding 50% overlapping.This delay generally can be carried out decorrelation to the such stationary signal in short-term of voice.Because voice signal also has correlation when long, i.e. the time-delay of 3 times of frame lengths sometimes still is not enough to remove fully its correlation, and this may cause the distortion of signal.Remove the correlation between s (n) and the y (n) for further raising voice quality need increase a de-correlation block, a lot of classical decorrelation methods are suggested, can be referring to document [2] (M.Ali; " Stereophonic acoustic echo cancellation system using time-varying all-pass filteringfor signal decorrelation, " in Proceedings of the ICASSP ' 98, vol.6; Pp.3689-3692, Seattle, USA; May 1998) and document [3] ([3] J.Benesty; D.R.Morgan, and M.M.Sondhi, " A betterunderstanding and an improved solution to the specific problems of stereophonic acoustic echo cancellation; " IEEE Trans.Speech Audio Processing; Vol.6, pp.156-165, Mar.1998).Need to prove that choosing of de-correlation must be prerequisite not damage voice quality.

Step 7) is optional) output of step 6) is increased a gain module 420.

System gain can be regulated through dual mode; The one, regulate the gain of digital signal; Promptly in gain module 420,

applied gain g, also can realize through the gain of regulating analogue system such as power amplifier.

Should be noted that the inhibition method of uttering long and high-pitched sounds described in the invention can use multiple mode to realize, the for example combination of hardware, software or hardware and software.Hardware platform can be FPGA, PLD or other application-specific integrated circuit ASICs.Software platform comprises DSP, ARM or other microprocessors.The combination of software and hardware for example part of module realizes that with dsp software part of module such as FFT realize with hardware accelerator.

It should be noted last that above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is specified with reference to embodiment; Those of ordinary skill in the art is to be understood that; Technical scheme of the present invention is made amendment or is equal to replacement, do not break away from the spirit and the scope of technical scheme of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.

Claims (8)

1. inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation, these method concrete steps comprise:

Step 1): the signal x (n) that respectively microphone is gathered with raise produce device output signal y (n) carry out branch frame, windowing and Fourier transform obtain the microphone frequency domain signal X (i, k) with raise produce device export frequency-region signal Y (i, k);

Step 2): (i, k) (i k) calculates crosspower spectrum φ between microphone and the speaker output signal to the microphone frequency domain signal X that obtains according to described step 1) with loud speaker output frequency-region signal Y _Yx(i is k) with the power spectrum φ of speaker output signal _Yy(i, k); And with crosspower spectrum φ _Yx(i is k) divided by power spectrum φ _Yy(i k) obtains the estimation of feedback path transfer function amplitude spectrum

Step 3): using said step 2) to obtain the transfer function of the feedback path estimate of the amplitude spectrum

calculate the feedback signal amplitude spectrum of the current frame

Step 4): utilize the microphone signal amplitude spectrum | X (i; K) | and feedback signal amplitude spectrum

the calculating acquisition gain function G of described step 3) acquisition (i, k);

Step 5): (i k) comes that (i k) revises and realizes uttering long and high-pitched sounds inhibition to the microphone frequency domain signal X to the gain function G that obtains according to described step 4).

2. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 1 is characterized in that this method also further comprises:

Step 6): the revised microphone frequency-region signal to described step 5) obtains is handled the final output signal of acquisition, and output signal

is carried out decorrelation handle; With

Step 7): described step 6) process result is carried out gain-adjusted.

3. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 1 is characterized in that described step 2) concrete steps that obtain feedback path transfer function amplitude spectrum comprise:

21): adopt the level and smooth method of single order autoregression, obtain crosspower spectrum φ according to formula (1) _Yx(i, k);

φ _yx(i，k)=α _Cφ _yx(i-1，k)+(1-α _C)|Y ^*(i，k)X(i，k)| (1)

Wherein, α _CBe smoothing factor, 0≤α _C＜1;

22): adopt the level and smooth method of single order autoregression, obtain power spectrum φ according to formula (2) _Yy(i, k);

φ _yy(i，k)=α _Rφ _yy(i-1，k)+(1-α _R)Y ^*(i，k)Y(i，k) (2)

Wherein, α _RBe smoothing factor, 0≤α _R＜1;

23): utilize crosspower spectrum φ _Yx(i is k) with power spectrum φ _Yy(i, k) root pick formula (3) obtains feedback path transfer function amplitude spectrum

$\left|{\hat{F}}^{\′}(i,k)\right|=\frac{{\mathrm{\φ}}_{\mathrm{yx}}(i,k)}{{\mathrm{\φ}}_{\mathrm{yy}}(i,k)+\mathrm{\δ}}---\left(3\right).$

4. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 1; It is characterized in that described step 3) is calculated according to formula (4) and obtained present frame feedback signal amplitude spectrum

$\left|\hat{X}(i,k)\right|=\mathrm{\β}\left|{\hat{F}}^{\′}(i,k)\right|\left|X(i,k)\right|---\left(4\right)$

Wherein, β is an adjustable parameter, carries out this parameter of choose reasonable according to actual conditions.

5. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 1 is characterized in that the method for described step 4) calculated gains function comprises: based on priori signal feedback ratio method with based on check back signal feedback ratio method.

6. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 5 is characterized in that, described step 4) based on check back signal feedback ratio method according to formula (5) calculate to obtain gain function G (i, k);

$G(i,k)={\left[\frac{\max ({\left|X(i,k)\right|}^{\mathrm{\α}}-{\left|\hat{X}(i,k)\right|}^{\mathrm{\α}},0)}{{\left|X(i,k)\right|}^{\mathrm{\α}}}\right]}^{\frac{1}{\mathrm{\α}}}---\left(5\right)$

Wherein, α＞0.

7. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 1 is characterized in that, described step 5) according to formula (6) to the microphone frequency domain signal X (i, k) revise and realize uttering long and high-pitched sounds inhibition:

$\hat{S}(i,k)=X(i,k)G(i,k)---\left(6\right).$

8. the inhibition method of uttering long and high-pitched sounds based on the feedback signal spectrum estimation according to claim 2 is characterized in that, described step 7) adopts the gain of regulating digital signal or the gain of regulating the analogue system power amplifier.

Images