JP2010157806A - Howling prevention device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a howling prevention device which can perform estimation processing of a loop gain while holding a power consumption down. <P>SOLUTION: A microphone unit 1 performs an estimation of a loop gain by using a pseudo noise, a PN code, and calculates a delay time of a closed loop or a distance between the microphone unit 1 and a loudspeaker 3. The microphone unit 1 changes a frequency of obtaining correlation by a correlation calculating portion 20 according to the estimated loop gain, the delay time of the closed loop, or the distance between the microphone unit 1 and the loudspeaker 3. Also, the microphone unit 1 changes a frequency of estimating the loop gain, or a frequency of calculating the delay time and the distance according to the estimated loop gain, the delay time of the closed loop, or the distance between the microphone unit 1 and the loudspeaker 3. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a howling prevention device for preventing howling.

  Conventionally, various methods for preventing howling have been proposed (see, for example, Patent Document 1).

In the audio signal amplifier circuit described in Patent Document 1, when howling is detected, a low-pass filter for removing howling is operated to lower the gain of the audio signal amplifier circuit. The audio signal amplifier circuit checks whether or not howling is detected by stopping the operation of the low-pass filter at predetermined intervals. The audio signal amplification circuit gradually decreases the gain until no howling is detected. In the audio signal amplifier circuit, the howling is fixed to a gain that is no longer detected.
JP-A-7-15788

  In a general howling removal apparatus such as the audio signal amplification circuit described in Patent Document 1, since howling detection processing is always performed, power consumption increases. For this reason, in particular, a howling removing device such as a microphone driven by a battery has a problem that the battery is consumed in a short time. Then, an object of this invention is to provide the howling prevention apparatus which can suppress an electric power consumption, preventing howling reliably.

  The howling prevention apparatus of the present invention generates pseudo noise and estimates a loop gain based on the generated pseudo noise. Then, according to the estimated loop gain, the howling prevention device has a correlation calculation unit that correlates the pseudo noise included in the audio signal collected by the sound collection unit and the pseudo noise generated by the pseudo noise generation unit. Change the desired frequency. For example, when the estimated loop gain is small, howling is less likely to occur, so the frequency of obtaining the correlation is reduced. When the estimated loop gain is large, howling is more likely to occur, and the frequency of obtaining the correlation is increased. To do. Thereby, the howling prevention device changes the frequency for obtaining the correlation between the pseudo noise included in the audio signal picked up by the sound pickup means and the pseudo noise generated by the pseudo noise generation means, thereby calculating the correlation calculation means. Thus, howling can be prevented while reducing power consumption.

  In addition, the howling prevention apparatus of the present invention changes the frequency at which the correlation calculation means obtains the correlation in accordance with the delay time from the generation of the pseudo noise until the correlation calculation means calculates the correlation value. If the delay time is short, there is a high possibility of howling. Therefore, the frequency of obtaining the correlation is increased. If the delay time and the sound propagation distance are long, the possibility of howling is reduced. By reducing the frequency of obtaining the correlation, the number of operations of the correlation calculating means can be reduced and the power consumption can be suppressed.

  Furthermore, the howling prevention apparatus of the present invention changes the frequency of obtaining the correlation if the estimated loop gain and the calculated delay time do not change for a certain period. If the loop gain and delay time do not change for a certain period and the same value, the howling prevention device does not move and stays in the same place, and the audio signal gain is constant, so that howling occurs for a while. The likelihood of doing so is expected to be low. Therefore, it is possible to reduce the number of operations of the correlation calculation means and suppress power consumption.

  In addition, the howling prevention apparatus of the present invention changes at least one of the length of pseudo noise generated by the pseudo noise generation means, the generation interval, and the number of sequences according to the estimated loop gain. When the loop gain is low and the possibility of howling is low, the number of operations of the pseudo noise generation means is reduced by increasing the pseudo noise generation interval or reducing the number of pseudo noise sequences. Power consumption can be suppressed.

  The howling prevention device of the present invention changes the frequency of performing the correlation process depending on whether or not howling is likely to occur, so by reducing the number of processes when there is no need to frequently perform the correlation process, It is possible to suppress power consumption while reliably preventing howling.

  FIG. 1A is a block diagram showing the overall configuration of a loudspeaker, and FIG. 1B is a block diagram showing the configuration of a microphone (microphone unit) incorporating a howling prevention device of the present invention. In the description of this embodiment, unless otherwise specified, all audio signals are digital signals, and a configuration for performing A / D conversion and D / A conversion is omitted. Moreover, although illustration is abbreviate | omitted, the microphone unit 1 shall be driven with the battery as an example.

  As shown in FIG. 1A, the loudspeaker 10 includes a microphone unit 1, an amplifier device 2, and a speaker 3 (sound emitting means).

  The microphone unit 1 includes a microphone element 11 (input means), a calculation unit 5, a pseudo noise superimposing unit 7, and a control unit 9.

  The sound collected by the microphone element 11 is output to the arithmetic unit 5 and the pseudo noise superimposing unit 7. The pseudo noise superimposing unit 7 superimposes pseudo noise on the audio signal picked up by the microphone element 11 and outputs it to the subsequent amplifier device 2. The audio signal output from the microphone unit 1 is amplified by the amplifier device 2, is emitted from the speaker 3, propagates in the air, and is collected by the microphone unit 1. Thus, in the loudspeaker 10, a closed loop is formed between the microphone unit 1 and the speaker 3.

  Since a closed loop is formed as described above, the microphone unit 1 measures the delay time of the closed loop and estimates the loop gain in the calculation unit 5 in order to prevent the occurrence of howling. The control unit 9 prevents the howling from occurring by controlling the gain of the audio signal output from the pseudo noise superimposing unit 7 according to the closed loop delay time and the loop gain output from the calculation unit 5. Further, the control unit 9 correlates the audio signal picked up by the microphone unit 1 and the pseudo noise generated by the calculation unit 5 according to the closed loop delay time and the loop gain value counted by the calculation unit 5. Change the desired frequency.

  Next, a detailed configuration of the microphone unit 1 will be described. As shown in FIG. 1B, the calculation unit 5 of the microphone unit 1 includes an M-sequence generator 15, an HPF 19, a correlation calculation unit 20, a timer 21, a gain estimation unit 22, and a calculation unit 23. The pseudo noise superimposing unit 7 of the microphone unit 1 includes an LPF 12, an audio signal volume 13, a superimposing unit 14, an M sequence generator 15, an N-times oversampling unit 16, an HPF 17, and a pseudo noise volume 18. .

  First, the configuration and function of the pseudo noise superimposing unit 7 will be described.

  The audio signal picked up by the microphone element 11 is input to the LPF 12 of the pseudo noise superimposing unit 7 and the HPF 19 of the calculating unit 5.

  The LPF 12 cuts a high frequency (for example, 20 kHz or more which is a frequency band that is difficult for humans to hear) from the collected audio signal, and outputs it to the audio signal volume 13.

  The audio signal volume 13 outputs a signal collected by the microphone element 11 to the superposition unit 14 with a gain set by the control unit 9.

  The M-sequence generator 15 corresponds to pseudo noise generation means, periodically generates a signal with high autocorrelation such as a PN code (M sequence) as pseudo noise, and outputs the signal to the N-times oversampling unit 16. The M-sequence generator 15 generates a PN code when a trigger signal periodically output from the control unit 9 is input.

  The pseudo noise is not limited to the M series, and other random numbers such as a Gold series may be used. Further, the output period of the pseudo noise is the time (the impulse response in the acoustic transmission system) until the reflected wave (indirect wave) component falls below a predetermined level so that the gain estimation unit 22 described later can perform loop gain estimation processing. Longer than the convergence time).

  The N-times oversampling unit 16 oversamples the pseudo noise signal (PN code bit string) output from the M-sequence generator 15 with a sampling clock having a frequency N times the bit frequency, and outputs the result to the HPF 17. Oversampling by the N-times oversampling unit 16 is not essential in the present invention, but by performing oversampling, the temporal redundancy of pseudo noise increases and the accuracy of correlation calculation can be improved. Actually, the presence or absence of oversampling may be set according to the required accuracy and the code length of the pseudo noise.

  The HPF 17 cuts the low frequency range (for example, less than 20 kHz) of the signal input from the N-times oversampling unit 16.

  Note that the LPF 12 and the HPF 17 are not indispensable configurations in the present invention, but these configurations can improve the audibility. That is, since the HPF 17 cuts the low noise (human audible band) of the pseudo noise, even if the pseudo noise is emitted from the speaker 3, it becomes difficult to hear the pseudo noise and there is no sense of incongruity in hearing. . Further, the LPF 12 prevents the high-frequency pseudo-noise once input to the microphone from being output again to the amplification system, thereby suppressing the pseudo-noise loop phenomenon. When the LPF 12 and the HPF 17 are not provided, the pseudo noise loop phenomenon can be suppressed by subtracting the pseudo noise component from the audio signal picked up by the microphone element 11 and outputting the subtracted noise to the amplification system.

  The signal output from the HPF 17 is input to the pseudo noise volume 18. The pseudo noise volume 18 outputs the output signal of the HPF 17 to the superimposing unit 14 with the gain set by the control unit 9. The level of the pseudo noise may be a weak level that does not cause a sense of incongruity, but a level that can detect the peak value of the pseudo noise correlation is secured.

  The superimposing unit 14 superimposes the signal (pseudo noise) output from the pseudo noise volume 18 on the audio signal output from the audio signal volume 13 and outputs the superimposed signal to the amplifier device 2.

  Next, the configuration and function of the calculation unit 5 will be described.

  The control unit 9 outputs a trigger signal to the M-sequence generator 15 and the timer 21 at a predetermined cycle.

  When the trigger signal from the control unit 9 is input, the M-sequence generator 15 outputs the same pseudo noise as that output to the N-times oversampling unit 16 to the correlation calculation unit 20.

  When the trigger signal from the control unit 9 is input, the timer 21 starts time measurement and transmits a timer signal indicating the count time to the calculation unit 23.

  The microphone element 11 picks up sound containing pseudo noise and outputs this sound signal to the HPF 19. The HPF 19 cuts a low frequency (for example, less than 20 kHz) from the sound signal picked up by the microphone element 11 and outputs the cut signal to the correlation calculation unit 20.

  The correlation calculation unit 20 obtains a correlation between the pseudo noise input from the M-sequence generator 15 and the output signal of the HPF 19 (the voice signal collected by the microphone element 11). Since the M-sequence code has a very high autocorrelation, if the output signal of the HPF 19 includes the same M-sequence pseudo noise, the level of the correlation value increases. The correlation calculation unit 20 outputs the correlation value at that time to the gain estimation unit 22 and the calculation unit 23 at the timing when the high level correlation value is calculated (reception timing). Moreover, although mentioned later for details, the correlation calculation part 20 changes the frequency which calculates | requires a correlation according to the control signal from the control part 9. FIG.

  When the correlation value is input from the correlation calculation unit 20, the calculation unit 23 refers to the timer signal (time count value) from the timer 21 and obtains a time difference from the timing at which the pseudo noise is output to the reception timing. This time difference corresponds to the delay time of the closed loop. The calculation unit 23 outputs this delay time information to the control unit 9. Further, the calculation unit 23 may be set to multiply the above delay time by the speed of sound to calculate the distance from the speaker 3 to the microphone unit 1 and to output this distance information to the control unit 9.

  The gain estimation unit 22 corresponds to estimation means, estimates a loop gain based on the correlation value, and outputs information on the estimated loop gain to the control unit 9. The loop gain estimation method may be in various modes. For example, the loop gain is estimated in the following manner.

  FIG. 2 schematically shows the time axis characteristic of the correlation value. The gain estimating unit 22 acquires the absolute value | A | of the correlation peak detected first after outputting the PN code. The waveform around the timing at which the first correlation peak is detected can be regarded as a direct sound (direct wave) returned to the microphone unit 1. For example, the gain estimation unit 22 estimates the absolute value | A | as a loop gain.

  Further, the gain estimation unit 22 acquires a correlation peak that appears before a predetermined time elapses from the first correlation peak (that is, until the M-sequence generator 15 outputs the next PN code). The gain estimator 22 selects an absolute value | A |, an absolute value | B |, and an absolute value | C | of a correlation peak at a predetermined level L or higher from among the acquired correlation peaks. The delay time of the spatial sound emission system from the speaker 3 to the microphone unit 1 is longer than the delay time of the signal processing system from the microphone unit 1 to the speaker 3. For this reason, the waveform until a predetermined time elapses from the first correlation peak can be determined as a reflected wave such as a wall. The gain estimation unit 22 estimates | A | + | B | + | C | as a loop gain as a result of integrating the absolute values of the correlation peaks of the direct wave and the reflected wave.

  In the above method, since the output period of the pseudo noise is set longer than the convergence time of the impulse response in the acoustic transmission system, the dummy noise is output until the next pseudo noise is output after the pseudo noise is output. The silent section may be eliminated by outputting. By always outputting a noise sound, the pseudo noise becomes inconspicuous and there is no sense of incongruity in hearing.

  The loop gain estimated by the gain estimation unit 22 as described above is output to the control unit 9.

  The control unit 9 controls the audio signal volume 13 based on the information input from the gain estimation unit 22 and the calculation unit 23 to adjust the gain of the audio signal.

  Specifically, the control unit 9 adjusts the gain according to the estimated loop gain value. That is, the control unit 9 outputs a control signal for lowering the gain to the audio signal volume 13, assuming that the possibility of howling increases as the estimated loop gain value approaches the predetermined threshold th. Further, the control unit 9 outputs a control signal for increasing the gain to the audio signal volume 13 on the assumption that the possibility of occurrence of howling decreases as the estimated loop gain value departs from the predetermined threshold th. Thereby, howling can be prevented beforehand.

  The control unit 9 may adjust the gain based on the closed-loop delay time calculated by the calculation unit 23 or the distance between the microphone unit 1 and the speaker 3 in addition to the loop gain value. That is, the control unit 9 determines that the possibility of occurrence of howling increases as the closed-loop delay time calculated by the calculation unit 23 or the distance between the microphone unit 1 and the speaker 3 becomes shorter. Output to the volume 13 for use.

  The predetermined threshold th may be set by a user operation input or may be a specified value.

  The control unit 9 may be configured to display a warning on a display unit (not shown) when the value of the loop gain approaches the predetermined threshold value th and howling is about to occur. In this case, the user having the microphone unit 1 can prevent howling by moving away from the speaker 3 in response to the warning display.

  Note that the controller 9 may perform gain adjustment and / or warning display.

  Next, based on the information input from the gain estimation unit 22 and the calculation unit 23, the control unit 9 controls the correlation calculation unit 20 to change the frequency for obtaining the correlation. Specifically, the control unit 9 increases the frequency of obtaining the correlation as the estimated loop gain value approaches the predetermined threshold th, and the estimated loop gain value departs from the predetermined threshold th. Accordingly, the correlation calculation unit 20 is operated so as to intermittently obtain the correlation by decreasing the frequency of obtaining the correlation.

  Further, the control unit 9 may change the frequency at which the correlation calculation unit 20 obtains the correlation based on the closed loop delay time or the distance between the microphone unit 1 and the speaker 3 instead of the loop gain value. That is, the control unit 9 increases the frequency of obtaining the correlation as the closed loop delay time or distance calculated by the calculation unit 23 is shortened, and decreases the frequency of obtaining the correlation as the closed loop delay time or distance is increased, The correlation calculation unit 20 is controlled so as to obtain the correlation intermittently.

  The control unit 9 changes the frequency at which the correlation calculation unit 20 obtains the correlation based on both the loop gain value and the closed loop delay time or the distance between the microphone unit 1 and the speaker 3. May be.

  That is, (1) When the loop gain value is close to the predetermined threshold th and the delay time of the closed loop or the distance between the microphone unit 1 and the speaker 3 is short, the control unit 9 always operates the correlation calculation unit 20 And control to obtain the correlation. (2) When the loop gain value is away from the predetermined threshold th and the delay time of the closed loop or the distance between the microphone unit 1 and the speaker 3 is longer than the predetermined value, the control unit 9 According to the value, the correlation calculation unit 20 is operated intermittently so that the time during which the operation is stopped becomes longer. Also, (3) when the loop gain value is close to the predetermined threshold th and the closed loop delay time or the distance between the microphone unit 1 and the speaker 3 is long, or (4) the loop gain value is the predetermined threshold value. When the closed loop delay time or the distance between the microphone unit 1 and the speaker 3 is short apart from th, the control unit 9 shortens the time during which the correlation calculation unit 20 stops operating according to those values. Operate intermittently so that

  In the above (2) to (4), for example, as the loop gain value is closer to the threshold value, and as the delay time or distance is shorter, the time during which the correlation calculation unit 20 has stopped operating is determined. It is good to operate intermittently so as to shorten.

  In the above (1), when the loop gain value, the closed loop delay time, or the distance between the microphone unit 1 and the speaker 3 becomes smaller than a preset threshold value, the correlation calculation unit 20 always operates. Good to do.

  It should be noted that the time for which the correlation calculation unit 20 is operated intermittently may be obtained by conducting an experiment in advance. Further, when dividing into cases as described above, a plurality of threshold values are set, and these threshold values, loop gain value, closed loop delay time, or distance between the microphone unit 1 and the speaker 3 are set. The frequency for obtaining the correlation may be set in accordance with the magnitude relationship of.

  Further, the code length of the M sequence may be changed according to the frequency of estimating the loop gain. For example, when the frequency of estimating the loop gain is high, the M sequence length may be shortened, and when the frequency of estimating the loop gain is low, the M sequence length may be increased. When the frequency of estimating the loop gain is high, since the gain of the audio signal is large and the distance between the microphone unit 1 and the speaker 3 is short, the correlation of the pseudo noise can be reliably obtained even if the M sequence length is shortened. it can. Although the calculation time changes according to the distance, it is possible to follow the environmental change by changing the code length according to the distance in this way.

  Further, the generation interval of the PN code to be used may be changed according to the frequency of estimating the loop gain. For example, when the frequency of estimating the loop gain is high, the generation interval of the PN code is made dense, and when the frequency of estimating the loop gain is low, the generation interval of the PN code is preferably made sparse. Thereby, since a PN code can be generated as necessary, the loop gain can be reliably estimated.

  Further, the number of PN code sequences to be used may be changed according to the frequency of estimating the loop gain. For example, when the frequency of estimating the loop gain is high, a plurality (for example, three) of different PN codes (M-sequence codes) are sequentially output from the M-sequence generator 15 with the timing slightly shifted. . On the other hand, when the frequency of estimating the loop gain is low, one PN code is output from the M-sequence generator 15 at a predetermined timing. In this way, even when the frequency of estimating the loop gain is high, the loop gain can be estimated continuously in a short time, so that howling can be prevented with high accuracy. Even when the frequency of estimating the loop gain is low, the loop gain can be reliably estimated.

  If the loop gain calculated by the gain estimation unit 22 and the delay time or the sound propagation distance calculated by the calculation unit 23 do not change for a certain period, the control unit 9 sends a control signal to the correlation calculation unit 20. The frequency of obtaining the correlation may be reduced. In this case, the distance between the microphone unit 1 and the speaker 3 is constant and the sound volume is constant, and howling is unlikely to occur. Therefore, the power consumption can be suppressed by reducing the frequency with which the correlation calculation unit 20 obtains the correlation and reducing the number of operations of the correlation calculation unit 20.

  The calculation unit 5 and the pseudo noise superimposing unit 7 may be built in the amplifier device 2 instead of the microphone unit 1. Further, an adapter including the arithmetic unit 5 and the pseudo noise superimposing unit 7 is configured, a microphone is connected to the input unit (input interface) of the adapter, and a signal input from the input unit is input to the arithmetic unit 5 and the pseudo noise superimposing unit. 7 may be provided. In any case, it suffices if it is provided before the amplification system such as an amplifier device.

It is a block diagram which shows the function and structure of a howling prevention apparatus. This is a schematic diagram of the time-axis characteristics of correlation values.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Microphone unit, 11 ... Microphone element, 12 ... LPF, 13 ... Volume for audio | voice signal, 14 ... Superimposition part, 15 ... M series generator, 17, 19 ... HPF, 20 ... Correlation calculation part, 21 ... Timer, 22 ... gain estimation part, 23 ... calculation part

Claims (4)

An input means for receiving an input of an audio signal;
Pseudo noise generating means for generating pseudo noise;
Sound emitting means for superimposing the pseudo noise on the audio signal and emitting sound;
Correlation calculation means for obtaining a correlation between the voice signal picked up by the input means and the pseudo noise generated by the pseudo noise generation means;
An estimation means for estimating a loop gain based on the correlation value obtained by the correlation calculation means;
Control means for changing the frequency at which the correlation calculating means obtains the correlation according to the loop gain estimated by the estimating means;
Howling prevention device equipped with. A calculation means for calculating a delay time from when the pseudo noise is generated until the correlation calculation means calculates a correlation value;
The howling prevention apparatus according to claim 1, wherein the control unit changes a frequency at which the correlation calculation unit obtains a correlation according to the delay time calculated by the calculation unit.   3. The control unit according to claim 2, wherein the control unit changes the frequency at which the correlation calculation unit obtains the correlation unless the loop gain estimated by the estimation unit and the delay time calculated by the calculation unit change for a certain period of time. The howling prevention apparatus in any one.   The control unit changes at least one of a length of pseudo noise generated by the pseudo noise generation unit, a generation interval, or a number of sequences according to a loop gain estimated by the estimation unit. The howling prevention apparatus in any one of.

Images