CN102164329B - De-noising assembly and noise-eliminating method thereof - Google Patents

Abstract

A denoising component, which includes a receiving microphone, a loudspeaker, a signal processing unit that performs noise denoising processing on a noisy voice signal received by the receiving microphone and outputs a denoising voice signal, and amplifies the denoising voice signal And output a signal amplifying unit for amplifying the voice signal. Wherein, the noise canceling component is further provided with: a front microphone, a development board and an adder unit. The present invention also provides a noise reduction method of the noise reduction component according to the present invention. Compared with related technologies, the noise canceling system of the present invention has a better noise canceling effect, and reduces the limitation of the position of the noise canceling system relative to the human ear when used.

Description

Noise canceling component and noise canceling method thereof

【Technical field】

The invention relates to a noise canceling component and a noise canceling method thereof, in particular to a noise canceling system component used in portable mobile consumer products and a noise canceling method thereof.

【Background technique】

With the development of wireless communication, there are more and more mobile phone users around the world. The requirements of users for mobile phones are not only satisfied with calls, but also to be able to provide high-quality call effects. Especially the current development of mobile multimedia technology, the mobile phone Call quality is even more important.

Due to the existence of a large amount of environmental noise, the signal-to-noise ratio of the voice signals collected by the microphones of mobile phones and other communication equipment is generally not high enough, especially in high-noise environments such as street cars, it is necessary to increase the volume of the speaker in the communication equipment to hear clearly. Therefore, it is necessary to increase the gain of the input voice and improve the communication quality by means of voice enhancement. Therefore, the advantages and disadvantages of the design of the noise reduction components have become the focus of attention.

The noise cancellation component in the related art includes a receiving microphone, a speaker and a signal processing unit electrically connected to the receiving microphone and the speaker, and the signal processing unit includes a signal amplifying unit. The microphone receives the ambient noise voice signal and is processed by the signal processing unit to obtain a noise-eliminated voice signal, and the noise-eliminated voice signal is amplified by the signal amplifying unit and then sent out by the speaker. Since the products that people use using the noise canceling components are earmuffs or in-ears, the distance between them and the human ear is relatively fixed during use, and the noise canceling parameters of the signal processing unit are also fixed, and the signal amplifying unit The scaling parameter of is also a fixed value. The above-mentioned structure makes it possible for the products using the noise-canceling component of the related technology to have a better effect only in a specific position. The noise canceling component can also be applied on the mobile phone. When the noise canceling component of the related technology is used on the mobile phone, since each person uses the mobile phone to place the mobile phone on the side of the ear, the degree of fit is different, and the noise canceling component of the related technology is different. However, the parameters of the noise component are fixed, which results in that the noise cancellation effect of the mobile phone in the related art is up and down, and the noise cancellation effect has a positional limitation.

Therefore, it is necessary to propose a new denoising component to solve the above problems.

【Content of invention】

The technical problem to be solved by the present invention is to provide a noise canceling component with good noise canceling performance and can reduce the position limitation of the noise canceling effect.

According to above-mentioned technical problem, designed a kind of denoising component, its purpose is to realize like this: a kind of denoising component, it comprises receiving microphone, loudspeaker, carries out denoising processing and The signal processing unit that outputs the voice signal after denoising and the signal amplifying unit that amplifies the voice signal after denoising and outputs the amplified voice signal are characterized in that: the denoising component is also provided with:

The front microphone is used to receive the detection signal and output the detection feedback signal;

A development board, including a programmable processing unit that analyzes and processes the detection feedback signal output by the front microphone and outputs a DC voltage, and a signal generation unit that sends a detection signal;

The adder unit is used to superimpose the amplified voice signal and the detection signal, and input the superimposed voice signal to the loudspeaker.

The present invention also provides a noise elimination method using the above-mentioned noise elimination component, the method includes the following steps:

A denoising method, the method is applied to a denoising component, the denoising component includes a receiving microphone, a loudspeaker, performing denoising processing on a noisy voice signal received by the receiving microphone and outputting a denoising voice signal after signal processing unit and the signal amplifying unit that amplifies the voice signal after the denoising and outputs the amplified voice signal, the denoising component is also provided with a front microphone, a development board and an adder unit, wherein the front microphone is used to receive a synchronous response And output the detection feedback signal, the development board includes a programmable processing unit that analyzes and processes the detection feedback signal output by the front microphone and outputs a DC voltage and a signal generation unit that sends a detection signal, and the adder unit is used for Superposing the amplified voice signal and the detection signal, and inputting the superimposed voice signal to the loudspeaker, the method includes the following steps:

Step 001, set a number of test points, use the receiving microphone to receive an external noisy voice signal at any test point, the noisy voice signal is processed by the signal processing unit to obtain a noise-eliminated voice signal; After the voice signal passes through the signal amplifying unit, the amplified voice signal is obtained, and the amplification factor of the signal amplifying unit is changed by manually adjusting the DC voltage output by the programmable processing unit to make the amplified voice signal reach an ideal state, and at this time The DC voltage value corresponding to the amplification factor of the signal amplification unit, the so-called ideal state refers to monitoring the frequency response curve of the amplified voice signal with a tester, and changing the amplification of the signal amplification unit by continuously adjusting the DC voltage output by the programmable processing unit Multiple, so that the frequency response curve displayed on the tester is adjusted to the lowest position;

The signal generating unit of the development board sends out a fixed-frequency detection signal to the speaker, and the speaker sends a synchronous response after receiving the detection signal; utilizes the front microphone to receive the synchronous response and outputs a corresponding The detection feedback signal is sent to the programmable processing unit; using the programmable processing unit to perform FFT analysis processing on the detection feedback signal to obtain an FFT value;

Step 002, repeat step 001 at different test points to obtain several sets of FFT values and DC voltage values;

Step 003: Establish a coordinate system with the FFT value as the abscissa and the DC voltage as the ordinate, draw several groups of FFT values and corresponding DC voltage values in the coordinate system, and obtain the FFT value and DC voltage through the coordinate system The modulation relation formula;

Step 004, providing a program reader-writer, using the program reader-writer to write the modulation relation into the programmable processing unit of the development board;

Step 005, sending a detection signal with a fixed frequency through the signal generating unit of the development board and outputting it to the speaker, and the speaker sends a synchronous response after receiving the detection signal;

Step 006, using the front microphone to receive the synchronous response from the speaker and output a corresponding detection feedback signal to the programmable processing unit;

Step 007, using the programmable processing unit to perform FFT analysis and processing on the detection feedback signal and output a corresponding DC voltage;

Step 008: Use the receiving microphone to receive an external noisy speech signal and input it to the signal processing unit, the signal processing unit processes the noisy speech signal and outputs a denoised speech signal, and the denoised speech signal is processed by the signal amplifying unit The voice signal after noise is amplified and the amplified voice signal is output, and the amplified voice signal and the detection signal are superimposed by the adder unit and then input to the speaker.

Preferably, the detection signal is a 250Hz single-frequency pulse signal.

Preferably, the time interval for the front microphone to receive the synchronous response is equal to the time interval for the signal generation unit to send the detection signal, and the duration for the front microphone to receive the synchronous response is the signal generation unit unit emits the probe signal twice for the duration.

Preferably, the time interval for the signal generating unit to send out the detection signal is 1s, and the duration for each time the detection signal is sent is 10ms; the time interval for the front microphone to receive the synchronous response is 1s, and each time The duration of receiving the detection signal is 20ms.

Preferably, 7 test points are set, and each test point is tested 7 times.

Compared with the related technology, the noise canceling system of the present invention has better noise canceling effect, and reduces the limitation of the fit of the noise canceling system to the human ear when the noise canceling system is in use.

【Description of drawings】

Fig. 1 is a block diagram of the design method of the noise canceling component of the present invention.

Fig. 2 is the FFT analysis test diagram of the noise canceling component of the present invention, wherein the abscissa is the FFT value, and the ordinate is the DC voltage value required to control the amplification factor of the signal amplifying unit.

【detailed description】

The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

When products using technologies related to noise-canceling components are used, they will produce a coupling effect when they are attached to the human ear. The greater the pressure, the tighter the fit, the louder the sound received by the human ear; the smaller the pressure, the more The looser the closure, the smaller the sound received by the human ear. However, various parameters of the related art noise canceling component are fixed, therefore, the related art noise canceling component can have a better noise canceling effect only at a specific position. The present invention has designed a kind of denoising assembly for above-mentioned problem promptly, and it can adjust the output gain of the loudspeaker by testing the tightness of fitting between the product using the technology of the present invention and the human ear, that is, to control the amplification factor of the voice signal output, When the product of the related technology is used, the noise canceling effect can be adjusted to an ideal state according to the different positions of the product and the human ear. details as follows:

As shown in Figure 1, a kind of denoising component 10, it comprises receiving microphone 1, loudspeaker 5, the signal processing unit 3 that will receive the noisy voice signal that microphone 1 receives and output denoising processing and the voice signal after denoising and will The voice signal is amplified after denoising and output to the signal amplifying unit 4 for amplifying the voice signal. Wherein, the denoising assembly 10 is also provided with:

The front microphone 2 is used to receive a synchronous response and output a detection feedback signal;

The development board 6 includes a programmable processing unit 62 that analyzes and processes the detection feedback signal output by the front microphone 2 and outputs a DC voltage, and a signal generation unit 61 that sends a detection signal;

The adder unit 7 is configured to superimpose the amplified voice signal and the detection signal, and output the superimposed voice signal for input to the speaker 5 .

The present invention also provides a noise reduction method of the noise reduction component 10 according to the present invention, the method includes the following steps:

Step 001, set a number of test points, utilize the receiving microphone 1 to receive the noisy voice signal of the outside world at any test point, the noisy voice signal is processed by the signal processing unit 4 to obtain the voice signal after denoising; the voice signal after denoising After passing through the signal amplifying unit 4, the amplified voice signal is obtained, and the amplification factor of the signal amplifying unit 4 is changed by manually adjusting the DC voltage output by the programmable processing unit 62 so that the amplified voice signal reaches an ideal state, and the signal amplifying unit 4 is obtained at this time. The DC voltage value corresponding to the magnification.

The signal generation unit 61 of the development board 6 sends a detection signal of a fixed frequency and outputs it to the speaker 5, and the speaker 5 sends a synchronous response after receiving the detection signal; the front microphone 2 is used to receive the synchronous response and output a corresponding detection feedback signal to the programmable processing unit 62; using the programmable processing unit 62 to perform FFT analysis processing on the detection feedback signal to obtain an FFT value. Among them, FFT is fast Fourier transform.

The so-called ideal state refers to the minimum frequency response curve of the amplified voice signal in this embodiment, that is, the frequency response curve of the amplified voice signal is monitored by a tester, and the signal is changed by continuously adjusting the DC voltage output by the programmable processing unit 62. The magnification of the amplifying unit 4 makes the frequency response curve displayed on the tester adjusted to the lowest position, which means that the amplified voice signal has reached an ideal state.

Step 002, repeat step 001 at different test points to obtain several sets of FFT values and DC voltage values. Different test points are set according to the principle of distance variation between the receiving microphone 1 and the sound source, that is, the distance between the receiving microphone 1 and the sound source at each test point is different.

Step 003: Establish a coordinate system with the FFT value as the abscissa and the DC voltage as the ordinate, draw several groups of FFT values and corresponding DC voltage values in the coordinate system, and obtain the FFT value and DC voltage through the coordinate system modulation relationship.

In this embodiment, specifically, the FFT value is set to x, the DC voltage value is set to y, the FFT value is used as the abscissa, and the DC voltage value is used as the ordinate to establish a coordinate system, and several groups of FFT values and corresponding The DC voltage values are plotted in the coordinate system and connected end to end in turn, and the general relationship between the corresponding FFT value and the DC voltage is obtained according to the connection diagram obtained in the coordinate system.

In this embodiment, 7 test points are taken for analysis, which are A1-A7 respectively. The 7 test points are all within the normal working range of the noise canceling component 10 . In order to improve the accuracy, this embodiment does 7 tests to each test point, and then takes the average value. The data measured by the test points are as follows:

Table I

The FFT average value in the above table is the average value after removing the maximum and minimum values from the 7 test data of each test point.

As shown in Figure 2, the connection diagram is obtained from the figure. The connection diagram described in this embodiment can be approximated as two sections of straight lines, A1, A2 and A3 are connected to form the first straight line, A4, A5, A6 and A7 are connected to form the second straight line, and each section of straight line can use the straight line equation That is, the general relationship between the FFT value and the DC voltage is obtained as follows:

y=a*x+b (1)

Among them, a is the amplification factor, and b is the correction factor.

Of course, the established coordinate system is different, and when the test points are different, the general relational expression obtained is not linear, and may be other expressions, but the principle is the same.

Substituting the FFT values obtained by any two test points and the corresponding DC voltage values into the general relational formula (1), the amplification factor a and the correction factor b are obtained, and the amplification factor a and the correction factor b obtained are substituted into The general relation (1) leads to a modulation relation.

Considering that the normal use of the noise canceling component 10 of the present invention is close to the human ear, the pressure range of the human on the noise canceling component 10 is 3-15N, and the range of the FFT value of the corresponding noise canceling speech signal is 50-300 , so the test is carried out within the normal working range of the noise canceling components.

In the present embodiment, two test points are respectively taken on the straight line of different sections and substituted into the general relational formula (1) to obtain the values of a and b:

$\left\{\begin{array}{ccc}\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.00106523</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0.4962</span>}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 155.7696</span>}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 300</span>}<span\; class="notranslate">\; )</span>\\ \mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.0029293</span>}<span\; class="notranslate">\; ,</span>& \mathrm{<span\; class="notranslate">\; b</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0.78639</span>}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 66.02</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 155.696</span>}<span\; class="notranslate">\; )</span>\end{array}\right.$

With a, the value of b is substituted for described general relation (1) to obtain modulation relation (2), as follows:

$\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.00106523</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.4962</span>}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 155.696</span>}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 300</span>}<span\; class="notranslate">\; )</span>\\ <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.0029293</span>}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 0.78639</span>}& <span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 66.02</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 155.696</span>}<span\; class="notranslate">\; )</span>\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

In fact, due to human error and system error, when the test points are selected differently, the tested data will change, but these are all allowable errors and will not affect the performance of the noise canceling system. The principle is the same.

Step 004, providing a program reader/writer (not shown), and using the program reader/writer to write the modulation relational expression into the programmable processing unit 62 of the development board 6 . Wherein, the program reader can be a computer or other devices that can be programmed to read and write programs.

Step 005, the signal generating unit 61 of the development board 6 sends out a detection signal with a fixed frequency and outputs it to the speaker 5, and the speaker 5 sends a synchronous response after receiving the detection signal.

Step 006, using the front microphone 2 to receive the synchronous response from the speaker 5 and output the corresponding detection feedback signal to the programmable processing unit 62;

Step 007, use the programmable processing unit 62 to perform FFT analysis and processing on the detection feedback signal and output a corresponding DC voltage. Specifically, the programmable processing unit 62 calculates and processes the DC voltage corresponding to the amplification factor of the signal amplification unit 4 according to the modulation relational formula (2) written therein, and then adjusts the amplification factor of the signal amplification unit 4 . In fact, in the noise canceling component 10 of the present invention, the adjustment of the amplification factor of the signal amplification unit 4 by the development board 6 is an instant dynamic process, that is, when the detection feedback signal changes continuously, the DC voltage output by the development board 4 after processing It is also constantly changing, and thus the magnification is constantly changing.

Step 008, use the receiving microphone 1 to receive the external noisy voice signal and input it to the signal processing unit 3, the signal processing unit 3 processes the noisy voice signal and outputs the denoised voice signal, and the denoised voice signal is processed by the signal amplifying unit 4. The voice signal is amplified and the amplified voice signal is output. At this time, the amplified voice signal is amplified by adjusting the magnification factor. The amplified speech signal and the detection signal are superimposed by the adder unit 7 and then input to the speaker 5 .

In this embodiment, the detection signal sent by the signal generating unit 61 is a 250 Hz single-frequency pulse signal. Because the use of low-frequency signals as detection signals will optimize the noise cancellation effect, and more importantly, the human ear is not sensitive to low-frequency signals such as 250Hz, which is not easy for people to detect and avoids audiovisual illusions. Of course, the frequency of the detection signal is not limited to 250 Hz, and other frequency values are also possible, but it is best to use a low frequency band.

In order to accurately control the amplification factor of the signal amplifying unit 4 to the noise-cancelled voice signal, the time interval for receiving the synchronous response by the front microphone 2 in this embodiment is equal to the time interval for sending the detection signal by the signal generating unit 61. The duration for the microphone 2 to receive the synchronous response is twice the duration for the signal generating unit 61 to send out the detection signal. More preferably, the time interval for the signal generating unit 61 to send the detection signal is 1s, and the duration for sending the detection signal each time is 10ms; the time interval for the front microphone 2 to receive the synchronous response is 1s, and each time it receives The duration of the synchronous response is 20ms. In fact, the time interval between receiving the synchronous response and sending out the detection signal is not only 1s, but can be other values, such as 2s. In order to ensure the accuracy of noise cancellation, the most suitable value can be between 1-10s. The duration for the front microphone 2 to receive the synchronous response can also be other values, and the most suitable value can be between 10-20ms.

Compared with the related technology, the amplification parameters of the signal amplifying unit in the noise canceling component of the present invention are set to be adjustable, and a development board for controlling the amplification factor of the signal amplifying unit is added, and the product of the related technology is measured by the development board when it is used. After the tightness of the fit of the human ear, a DC voltage is sent to control the amplification factor of the signal amplification unit, so that the amplification factor can be adjusted in real time, so that the noise cancellation effect of related technical products is better.

What has been described above is only the embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art, improvements can be made without departing from the creative concept of the present invention, but these all belong to the present invention. scope of protection.

Claims (5)

1. a noise-eliminating method, it is characterised in that: the method is applied to de-noising assembly, described de-noising group Part includes receiving mike, speaker, being carried out by the Noisy Speech Signal that described reception mike receives Denoising Processing and after exporting de-noising voice signal signal processing unit and by voice signal after described de-noising Being amplified and export the signal amplification unit amplifying voice signal, described de-noising assembly is additionally provided with preposition Mike, development board and adder unit, wherein before microphone be used for receiving sync response defeated Going out and detect feedback signal, described development board includes the detection feedback signal to the output of described front microphone It is analyzed the processing unit able to programme processing and exporting DC voltage to send out with the signal sending detectable signal Raw unit, described adder unit is folded for described amplification voice signal and detectable signal are carried out signal Adding, voice signal after superposition is input to described speaker, the method comprises the steps:

Step 001, set some test points, utilize described reception mike to receive at any test point Extraneous Noisy Speech Signal, described Noisy Speech Signal processes through described signal processing unit Voice signal after de-noising；After de-noising, voice signal is amplified language after described signal amplification unit Tone signal, manually regulates the DC voltage of described processing unit able to programme output and changes described letter The amplification of number amplifying unit makes described amplification voice signal reach perfect condition, obtains now described The DC voltage value that the amplification of signal amplification unit is corresponding, so-called perfect condition refers to use tester The frequency response curve of voice signal is amplified in monitoring, by not stopping to regulate the direct current of processing unit able to programme output Voltage and change the amplification of signal amplification unit, make display frequency response curve on a tester be transferred to Extreme lower position；

The detectable signal being sent fixed frequency by the signal generating unit of described development board is exported to described Speaker, described speaker sends sync response after receiving described detectable signal；Utilize described preposition The mike described sync response of reception the corresponding detection of output feed back signal to described process list able to programme Unit；Utilize described processing unit able to programme that described detection feedback signal is made fft analysis process to obtain FFT value；

Step 002, repeat step 001 in different test point, obtain some groups of FFT values and unidirectional current Pressure value；

Step 003, with FFT value as abscissa, DC voltage value is that vertical coordinate sets up coordinate system, will Some groups of FFT values are drawn in a coordinate system with corresponding DC voltage value, are drawn by described coordinate system FFT value and the modulation relational expression of DC voltage；

Step 004, provide a program read write line, utilize described program read write line by described modulation relation Formula writes the processing unit able to programme of described development board；

Step 005, signal generating unit by described development board send the detectable signal of fixed frequency Output is to described speaker, and described speaker sends sync response after receiving described detectable signal；

Step 006, described front microphone is utilized to receive the described sync response that sends of described speaker And the corresponding detection of output feeds back signal to described processing unit able to programme；

Step 007, utilize described processing unit able to programme that described detection feedback signal is made fft analysis Process and export the DC voltage of correspondence；

Step 008, described reception mike is utilized to receive extraneous Noisy Speech Signal and be input to described Signal processing unit, described signal processing unit exports voice after de-noising after processing Noisy Speech Signal Signal, is amplified voice signal after described de-noising by signal amplification unit and exports amplification voice Signal, utilizes described adder unit to input to after described amplification voice signal and detectable signal superposition Described speaker.

Noise-eliminating method the most according to claim 1, it is characterised in that: described detectable signal is 250Hz pure-tone pulse signal.

Noise-eliminating method the most according to claim 2, it is characterised in that: described front microphone connects Receive the time interval of described sync response and described signal generating unit and send the time of described detectable signal Being spaced equal, the persistent period of the described front microphone described sync response of reception is that described signal occurs Unit sends the twice of the persistent period of described detectable signal.

Noise-eliminating method the most according to claim 3, it is characterised in that: described signal generating unit The time interval sending described detectable signal is 1s, and the persistent period every time sending described detectable signal is 10ms；It is 1s that described front microphone receives the time interval of described sync response, receives described every time The persistent period of sync response is 20ms.

Noise-eliminating method the most according to claim 4, it is characterised in that: described test point arranges 7 Individual, each described test point is tested 7 times.