TWM626673U - Device for determining direction of acoustic source - Google Patents

Abstract

The present invention is a sound source localization device for detecting the original azimuth angle of an original sound source relative to the sound source localization device, which comprises: a top cover provided with one or more air circulation holes, and its The distribution length has an inlet diameter; a base is provided with an accommodating space and is designed to be closely combined with the top cover; a microphone circuit board has a plurality of microphones on it, which can generate acoustic and electrical signals under the change of air pressure, and communicate with the one or a plurality of air circulation holes are placed in the accommodating space at a vertical distance, wherein the vertical distance is at least three times greater than the diameter of the inlet; and an arithmetic unit is placed in the accommodating space and electrically connected to the microphone circuit board The connection is used for receiving the acoustic and electrical signals, detecting the intensity of the acoustic and electrical signals, and outputting the original azimuth at the center of symmetry between the azimuths of the microphones according to the intensity of the acoustic and electrical signals. Wherein, the maximum distance between the microphones is greater than or equal to 1.3 times the vertical distance.

Description

Sound source localization device

This creation relates to a sound source localization device; especially a sound source localization device utilizing the Doppler effect, mainly in its high adaptability.

With the development of the semiconductor industry, microphones have changed from dynamic microphone/moving coil microphone, which used magnetic force, to electret condenser microphone/ECM, and even the newly emerging micro-electromechanical type. Microphone (MEMS microphone), the volume and power consumption are decreasing step by step. The ever-shrinking microphone mechanism can even be integrated with the amplifier or signal processor on the same circuit board, which pushes the microphone to move away from the passive sound-receiving role and step into the active operating interface.

Therefore, whether it is the Amazon Echo launched by Amazon in the United States or the Google Home launched by Google, an array of micro-electromechanical microphones is used to actively detect the direction of the speaker, thereby enhancing the recognition of voice commands. According to the principle of the angle of arrival (AoA) of the sound wave to form a phase difference between the microphones, by comparing the time difference of the acoustic and electrical signals generated by any two microphones, the speaker can be obtained from the connection of the two microphones with the largest time difference. azimuth.

If the speed of sound at 20℃ is 34,300cm/s, the mutual delay of 1ms represents the sound path difference of 34.3cm, Lm=Dm. cos(A), where Dm is the distance between the two microphones and A is the angle of arrival (see Figure 1). However, home devices like the Amazon Echo or Google Home have a maximum diameter of only At about 7cm, a mutual delay analysis capability of at least 0.2ms accuracy is required to find the azimuth of the speaker.

More importantly, in order to be basically dustproof and waterproof, the microphones are mostly placed inside the device, and the small holes used for sound collection are much smaller than the human voice wavelength range of 10-110cm. For this type of device, the sound wave passes through the small hole of the sound receiver in the form of a new point source, and at the same time loses the propagation direction of the original sound source, so the conventional single sound source plane wave model can no longer be applied.

Introduce a sound source localization method, which is implemented by the sound source localization device of the present creation, at least by the position of a microphone array relative to an entrance and the intensity of a plurality of acoustic and electrical signals generated by the microphone array to determine the position of an original sound source relative to the entrance. The original azimuth, which includes: determining one or more polar angles and a complex azimuth of the microphone array in the plane direction of the entrance; from the complex azimuth and the polar angle within a selected range The intensity of the complex acoustic and electrical signals determines a set of intensity distributions; determines a center of symmetry according to the set of intensity distributions; and determines that the original azimuth is located on the center of symmetry and deviates from the weaker side of the set of intensity distributions, wherein the selected range is from Select within 33° or more and less than 90°.

35°，該選定範圍係自大於等於50°以及小於90°內選擇，且隨著所述到達角愈趨向低角度，該選定範圍則愈趨向高角度；以及當該原始聲源相對於該入口之到達角

65°，該選定範圍係自大於等於33°以及小於等於63°選擇，且隨著所述到達角愈趨向高角度，該選定範圍則愈趨向低角度。 In the practical application of this creation, when the arrival angle of the original sound source relative to the entrance is not particularly limited, the selected range is selected from greater than or equal to 53° and less than 90°; when the original sound source arrives at the entrance relative to the entrance horn

35°, the selected range is selected from greater than or equal to 50° and less than 90°, and as the angle of arrival becomes lower, the selected range becomes higher; and when the original sound source is relative to the inlet angle of arrival

65°, the selected range is selected from greater than or equal to 33° and less than or equal to 63°, and as the angle of arrival tends to a higher angle, the selected range tends to a lower angle.

In one embodiment, the meaning of determining a center of symmetry according to the set of intensity distributions is to use the set of intensity distributions as weights to make a weighted average, and determine the center of symmetry accordingly.

In one embodiment, the meaning of determining a center of symmetry according to the set of intensity distributions includes: determining a silent region of the set of intensity distributions according to the noise level of the complex acoustic and electrical signals; and Find out the intensity distribution ratios at both ends of the outer edge of the silent area, and determine the center of symmetry accordingly.

In one embodiment, the meaning of determining a center of symmetry according to the set of intensity distributions includes: determining two extremes of the set of intensity distributions and a valley between the two extremes, wherein the two extremes and the valley The drop is greater than a threshold, otherwise, it is determined that the determination of the two extremes fails; when the determination of the two extremes is successful, it is determined that the symmetry center is located in the valley; and when the determination of the two extremes fails, according to the above The complex azimuth and the polar angle within a final range determine a set of final distributions from the complex acoustic and electrical signal strength, and a center of symmetry is determined according to the set of final distributions, wherein the minimum angle in the final range is greater than the selected The minimum angle in the range, and the threshold value is greater than or equal to one-fifth of the intensity of the diode value.

A sound source locating device for detecting the original azimuth angle of an original sound source relative to the sound source locating device, which comprises: a top cover, which is provided with one or more air circulation holes, and its distribution length has a The inlet diameter; a base with an accommodating space and designed to be closely combined with the top cover; a microphone circuit board with a plurality of microphones on it, which can be subjected to changes in air pressure to generate acoustic and electrical signals, and communicate with the one or more microphones The air circulation hole is placed in the accommodating space at a vertical distance, wherein the vertical distance is at least three times greater than the diameter of the inlet; and an arithmetic unit is placed in the accommodating space and electrically connected to the microphone circuit board, using to receive the acousto-electric signals, detect the intensity of the acousto-electric signals, and determine the azimuth angle between the microphones according to the intensity of the acousto-electric signals The center of symmetry outputs the original azimuth angle, wherein the maximum distance between the microphones is greater than or equal to 1.3 times the vertical distance. In order to obtain the original azimuth angle accuracy in all directions, the maximum distance between the microphones is greater than or equal to 2.38 times the vertical distance; or to determine the original sound source from directly in front, the maximum distance between the microphones is less than or equal to the vertical distance. 3.92 times the distance.

A sound source locating device for detecting the original azimuth angle of an original sound source relative to the sound source locating device, which comprises: a top cover with a sliding cavity, the inner side of which is provided with sliding tracks; an entrance A part, placed in the sliding cavity, can be manually or electrically actuated to move on the sliding track, and has one or more through-channels for air circulation, wherein the one or more through-channels have an inlet distributed along the length Diameter; a base with an accommodating space and designed to be closely combined with the top cover; a microphone circuit board with a plurality of microphones on it, which can be subjected to changes in air pressure to generate acoustic and electrical signals and communicate with the one or more microphones. The track is placed in the accommodating space at a vertical distance; and an arithmetic unit is placed in the accommodating space and electrically connected to the microphone circuit board for receiving the acoustic and electrical signals and detecting the intensity of the acoustic and electrical signals , and output the original azimuth angle at the center of symmetry between the microphone azimuth angles according to the strength of the acoustic and electrical signals, wherein, the movement of the entrance part on the sliding track will cause the vertical distance to increase or decrease, but the entrance part must be fixed at There are at least two places on the sliding track, and the vertical distance corresponding to the two places has a minimum value and a maximum value, and the minimum value is at least three times larger than the inlet diameter.

In one embodiment, the sound source localization device further includes: a motor electrically connected to the sliding track and placed in the accommodating space for adjusting the height of the entrance part relative to the sliding track by a control signal. The control signal is a periodic pulse signal which alternately appears to adjust the height of the inlet part. According to the time when the control signal appears alternately, when the intensity distribution of the acoustic and electrical signals has two extreme values, the computing unit outputs the result when the inlet component is higher, On the contrary, the result when the entry part is lower is output.

A: Arrival angle

A ₁ : First Arrival Angle

A ₂ : First Arrival Angle

B: Shell

Bin: lumen

C30: The path of α=30° on the plane of the microphone array

C60: α=60° path on the plane of the microphone array

C90: α=90° path on the plane of the microphone array

C120: α=120° path on the plane of the microphone array

C150: α=30° path on the plane of the microphone array

COM: Path surrounded by the same polar angle on the plane of the microphone array

c: speed of sound

Dm: distance between two microphones

D ₁ : Distribution width

D ₂ : Symmetric circle diameter

dp: depth

dur: duration

Lm: The sound path difference caused by the distance between the two microphones

L ₁ : The first sound path difference

M: Microphone circuit board

MU: arithmetic unit

O ₁ : The first entrance

O ₂ : Second entrance

OM: Propagation Displacement

p: walking time

SO: Incident direction

th: vertical distance

U: top cover

UB: base

V ₁ : first speed

α: interior angle

β: projection angle

γ: Doppler factor

δ: Instantaneous width of equivalent sound source

θ: polar angle where the microphone is located

Φ: The azimuth where the microphone is located

1: The time pulse from the left point source to the directly below

2: The right point source is transmitted to the time pulse directly below

3: The time pulse from the left point source to the next door

4: The time pulse from the right point source to the next door

335: Intensity distribution on COM at θ=30° incident at A ₁ = 35°

355: Intensity distribution on COM at θ=30° incident at A ₁ = 55°

375: Intensity distribution on COM with A ₁ = 75° incident at θ = 30°

435: Intensity distribution on COM at θ=45° incident at A ₁ = 35°

455: Intensity distribution on COM at θ=45° incident at A ₁ = 55°

475: Intensity distribution on COM at θ=45° incident at A ₁ = 75°

735: Intensity distribution on COM at θ=70° incident at A ₁ = 35°

755: Intensity distribution on COM at θ=70° incident at A ₁ = 55°

775: Intensity distribution on COM at θ=70° incident at A ₁ = 75°

90: Activity Pillar

901: Upper surface

902: Sliding sides

95: Display unit

Figure 1 is a schematic diagram of the plane wave model of the prior art

Figure 2 is a schematic diagram of the predicament of the prior art encountering two ventilation holes

Figure 3 is a schematic diagram of the configuration of the first embodiment of the present creation

Figure 4 shows the intensity distribution of acoustic and electrical signals in the first embodiment of the present invention

Figure 5 shows the intensity distribution of acoustic and electrical signals in the second and third embodiments of this creation

Figure 6 is a schematic diagram of the errors of the second and third embodiments of this creation

Figure 7 is a schematic diagram of the appearance of the fourth embodiment of the present invention

In order to illustrate the unique effect of this creation, the first two paragraphs first demonstrate the difficulties faced by conventional techniques with the use of conventional techniques and the configuration of the two radio holes; then, an embodiment of this creation is presented. Since this specification uses the acoustic and electrical signals sensed inside the device as an example, the subsequent analysis results of the mutual delay between the acoustic and electrical signals cannot be equivalent to dividing the sound path difference mentioned in the prior art by speed of sound.

Please refer to "Picture 2", when the original sound source from the left touches the two sound-receiving holes before and after, because the pores of the second sound-receiving holes are much smaller than the wavelength of human voice, after the sound wave passes through the two sound-receiving holes, it can no longer be Continue to propagate to the microphone in the original direction of travel, but instead radiate outward in the form of a new point source. The sound path difference caused by the traveling direction of the original sound source and the angle of arrival between the two radio holes along the line is c. p, where c is the speed of sound at that time, and p is the traveling time, which represents the time difference between the left point source and the right point source of the newly-born left point source of the two radio holes. The two radio holes The two microphones below, in addition to receiving sound waves from directly above, will also receive sound waves from next door. Figures 1 and 4 represent the time when the left point source is transmitted directly below and the right point source is transmitted to the next door; at the same time, Figures 2 and 3 represent the right point source is transmitted directly below and the left point respectively. The source is transmitted to the time next door.

When the time of the left and right point sources to the next door is slightly less than one of the walking time p, the left point source is transmitted to the right point source earlier than the right point source because the left point source is one the walking time p earlier than the right point source. microphone below. As shown in the upper left of "Figure 2", the arrival time of the left microphone (solid line) and the arrival time of the right microphone (dotted line) are not consistent, so the waveforms of the acoustic and electrical signals generated by the two microphones are quite different; this is different from the coherence measurement. A prerequisite for coherent measurement (CM) or cross-power spectral phase (CSP) - "waveforms that are similar to each other", is quite confusing. Since the waveforms of the acoustic and electrical signals generated by the two microphones are not similar, the mutual delay analyzed by the prior art will be different from p, and the azimuth angle of the original sound source cannot be known.

Please refer to "Fig. 3", which is the first embodiment of the present invention. A casing B is provided with a first inlet O ₁ ; and the first inlet O ₁ is provided with one or more air through holes distributed along the xy plane, and along the The distribution width D ₁ of the x-axis is smaller than the minimum wavelength of human voice. A microphone array is disposed in the cavity Bin surrounded by the casing B, and the depth dp from the plane where the microphone array is located to the casing B is at least three times greater _than the distribution width D1. When the original sound source strikes from the left along an incident direction SO, that is, an equivalent sound source moving at the first velocity V ₁ is formed at the first inlet O ₁ , and V ₁ =D ₁ /(L ₁ /c)= c/cos(A ₁ ), wherein L ₁ is the first sound path difference, A ₁ is the first angle of arrival, and the direction of V ₁ is the same as the direction of the x-axis. If a pulse wave is used to indicate the occurrence of the equivalent point source, several pulse waves appear in sequence over time to indicate the movement of the equivalent sound source, and the period during which the pulse waves appear is called a duration. )dur. Since the depth dp is much larger than the distribution width D ₁ , the time for the equivalent sound source to reach the microphone array is only related to the length of the propagation displacement OM from the first inlet O ₁ to the microphone array, and the The azimuth of the original sound source.

However, the duration dur is clearly related to the _first angle of arrival A1. For the microphone whose propagation displacement OM deviates more from the first velocity V ₁ , the equivalent sound source that occurs earlier has a shorter propagation displacement OM, and the equivalent sound source that occurs later has a longer propagation displacement OM. the propagation displacement OM _of There is a shorter OM for this propagation displacement. This is the same as the Doppler effect, the result is that when a moving sound source is coming towards us, we hear the higher frequency (the equivalent sound source is densely present) sound; otherwise, we hear the lower frequency sound (the equivalent sound source appears separately).

The duration dur(α) is the relational formula that obeys the Doppler effect: dur(α)=δ+γ. (L ₁ /c)...(1), where α is the inner angle formed by the first velocity V ₁ and the propagation displacement OM, δ is the instantaneous width of the equivalent sound source and the Doppler factor γ= |1-cos(α)/cos(A ₁ )|. The instantaneous width δ of the equivalent sound source is related to the distribution width of the first entrance O ₁ along the y-axis, and is proportional to the sine function of the projection angle β between the propagation displacement OM and the y-axis; therefore, the microphone on the xz plane The propagation displacement OM corresponds to the shortest δ, and the propagation displacement OM of the microphone on the yz plane corresponds to the longest δ. According to the law of conservation of energy, the intensity of the sound source passing through the propagation displacement OM should be inversely proportional to the square of the length of the propagation displacement OM, and the intensity of the acoustic signal of the microphone on the propagation displacement OM should be inversely proportional to the duration dur.

Please refer to "Fig. 4", on the plane where the microphone array is located, the curves C30, C60, C90, C120 and C150 corresponding to the inner angles α=30°, 60°, 90°, 120° and 150°. If you follow an equal-length path COM (a circle surrounded by a polar angle θ=70°), take the acoustic and electrical signal strength of the microphone array, sort by the azimuth angle Φ of the microphone array (horizontal axis in the right figure), and set 0° is the azimuth angle represented by the first velocity V ₁ , and the symmetry between the intensities of the acoustic and electrical signals can be seen (as shown in the figure on the right). Squares, triangles, and circles indicate that when the _first angle of arrival A1 is 35°, 55°, and 75°, the intensity of the acoustic and electrical signals varies with the azimuth of the microphone array (735, 755, 775). Therefore, by taking the weighted average of the azimuth angle Φ where the microphone array is located, the estimated value of the azimuth angle of the first velocity V ₁ can be obtained. ° and 6.7°, the average error is about 0.09. Estimated at a distance of six meters, the maximum misjudgment range caused by this average error is about 55cm.

In addition, since the internal angle α is greater than 80°, the intensity of the acoustic and electrical signals will be so weak that it is merged into the background noise and cannot be judged; therefore, the silent area where the intensity of the acoustic and electrical signals cannot be judged can be used to determine the The estimated value of the azimuth angle of the _first velocity V1. Furthermore, the ratio of the intensity of the acoustic and electrical signals at the outer edge of the silent area can also be used to calculate the estimated value of the azimuth angle of the first velocity V ₁ , for example: the outer edge of the silent area is +100° and -100° respectively. With the acoustic and electrical signal strengths 1 and 0.6, then [(+100°.0.6)+(-100°.1)]/1.6=-25°, which is the estimated value of the azimuth angle of the _first velocity V1.

Please refer to "Fig. 5", the second embodiment and the third embodiment of the present invention are respectively along the equal-length paths (polar angles θ=45° and 30°) surrounded by different polar angles, taking the distance of the microphone array The acoustic and electrical signal strengths are sorted by the azimuth angle Φ where the microphone array is located (horizontal axis in the right figure), and 0° is set as the azimuth angle represented by the first velocity V ₁ . Squares, triangles and circles respectively indicate that when the _first angle of arrival A1 is 35°, 55° and 75°, the intensity of the acoustic and electrical signals varies with the azimuth angle of the microphone array; among them, the figures 435, 455 and 475 respectively represent the three microphones on the equal-length path surrounded by the first angle of arrival at the polar angle θ=45°, and the diagrams 335, 355 and 375 respectively represent the three types of the first angle of arrival at the polar angle θ=30 ° the microphones on the path of equal length. Similarly, taking the weighted average of the azimuth angle Φ where the microphone array is located with the strengths of the acousto-electric signals, the average azimuth angle error Φ of the first velocity V ₁ obtained is 0.10 (θ=45°) and 0.12, respectively. (θ=30°). Although, the average error performance of the second embodiment and the third embodiment of the present invention is not as good as that of the first embodiment; the second embodiment has a better azimuth angle analysis for the first angle of arrival A ₁ =55° The third embodiment has better azimuth resolution capability for the first angle of arrival A ₁ =75°, and the error ratio is lower than 0.05.

Therefore, if the first angle of arrival A1 has _a relatively concentrated range in this creative application, the position of the microphone array should be selected for the range. For example: when the _first angle of arrival A1 is concentrated around 72°, it is suitable to select the equal-length path surrounded by the polar angle θ around 35°-40° to arrange the microphone array; and when the first angle of arrival is around 35°-40° If A ₁ is concentrated around 20°, it is suitable to choose to arrange the microphone array on the equal-length path surrounded by the polar angle θ greater than or equal to 60°. In practical application, by adjusting the depth of the depth, for example, raising the circuit where the microphone array is located, the configuration optimized for the _first angle of arrival A1 can be achieved.

Please refer to "Fig. 6", if the distribution of the microphone array on the equal-length path (viewed by the polar angle θ=70°) is not the axis of symmetry of the first velocity V ₁ , for example: the above figure shows The result with a 10° offset and the following figure shows the result with a 30° offset; the maximum error will come to 0.19 of the first angle of arrival A ₁ =75° (the circular result with a 10° offset in the upper figure) , and the worst case of the other offsets is about 0.14 (the result of the triangular mark with a 30° offset in the figure below). If the concentrated range of the first angle of arrival A ₁ is approximately greater than 50° in application, and the microphone array is distributed on the path of equal length (θ=30°-45°), the estimated value of the first velocity V ₁ can be controlled On the contrary, by distributing the microphone array on the same length path (θ>45°), the error ratio of the estimated value of the first velocity V ₁ can be controlled not to exceed 0.09.

Please refer to "Fig. 7", the fourth embodiment of the present creation, a sound location identification device is provided with a top cover U facing upward, and a second inlet O ₂ is provided; and a base UB is provided with a accommodating space , the design is closely combined with the top cover U. The second inlet O ₂ is provided with one or more air through holes distributed approximately in a circular symmetry along the xy plane, and the diameter D ₂ of the symmetry circle is smaller than the minimum wavelength of human voice. The second inlet O ₂ is located on the upper surface 901 of a movable pillar 90 , and the second inlet O ₂ and a microphone circuit board are adjusted by sliding the sliding side 902 of the movable pillar 90 up and down relative to the top cover U. The vertical distance th of M, and the vertical distance th is at least three times greater than the diameter D ₂ of the circle of symmetry. There is also an arithmetic unit MU and a display unit 95 electrically connected to the listening and positioning device. The computing unit MU is electrically connected to the microphone circuit board M, and is used to process the acoustic and electrical signals generated by the microphone, analyze the azimuth symmetry represented by the strength of the acoustic and electrical signals, and finally output it to the display unit 95 , so that the user can clearly see the speaker's orientation on the outside, for example: turn on the indicator light on the speaker's direction.

In response to the use situation of the lower speaker position, the upper surface 901 can be pressed, or the second inlet O ₂ can be slid to a lower position by the motor operation, that is, the vertical distance th can be made smaller, so as to increase the size of the speaker. the value of the polar angle.

On the other hand, the listening position identification device is placed on a wall or on the slope of a podium, with the top cover U facing the direction of most of the speakers, and the speaker's sound waves contact the second entrance _O2 . The two arrival angles A ₂ are both greater than or equal to 65°; therefore, the polar angle θ surrounded by the plurality of microphones on the microphone circuit board M to the second inlet O ₂ is preferably between 33° and 63°. In response to such a situation, the upper surface 901 can bounce under pressure, or slide the second inlet O ₂ to a higher position through motor operation, that is, the vertical distance th is larger, so as to reduce the value of the polar angle .

The audio signal mentioned in this manual is the result of the microphone's response to changes in air pressure. The generated power-related signal is not limited to its digital or analog form; and the acousto-electric signal strength can refer to the amplitude of the first up and down waveform in the acousto-electric signal, or it can refer to the highest amplitude of the acousto-electric signal within a time range. The value can be the energy density of the frequency spectrum of the acoustic signal after the fast Fourier transformation is completed, for example, the specific frequency is the lowest or second lowest main frequency of the human voice.

In the fifth embodiment of the present invention, an inner microphone array (θ=33°) and an outer microphone array (θ=68°) are arranged on a circle surrounded by two different polar angles, and their acoustic and electrical signals are analyzed separately. The inner and outer azimuths where the symmetry axis of the intensity is located. Next, it is detected whether there are two obvious extreme values in the acoustic and electrical signal, and when the two obvious extreme values exist, the inner azimuth angle is used as the result; otherwise, the outer azimuth angle is used as the result.

To sum up, the sound source localization method and device of this creation have indeed met the requirements of the patent application, and a patent application can be filed in accordance with the law. However, the above are only the preferred embodiments of this creation, and should not limit the scope of implementation of this creation; therefore, any simple equivalent changes and modifications made according to the scope of the patent application for this creation and the contents of the description are all applicable. shall remain within the scope of this creative patent.

A ₁ : First Arrival Angle

B: Shell

Bin: lumen

D ₁ : Distribution width

dp: depth

dur: duration

O ₁ : The first entrance

OM: Propagation Displacement

SO: Incident direction

V ₁ : first speed

α: interior angle

β: projection angle

θ: polar angle where the microphone is located

Claims (5)

A sound source locating device for detecting the original azimuth angle of an original sound source relative to the sound source locating device, which comprises: a top cover, which is provided with one or more air circulation holes, and its distribution length has a The inlet diameter; a base with an accommodating space and designed to be closely combined with the top cover; a microphone circuit board with a plurality of microphones on it, which can be subjected to changes in air pressure to generate acoustic and electrical signals, and communicate with the one or more microphones The air circulation hole is placed in the accommodating space at a vertical distance, wherein the vertical distance is at least three times greater than the diameter of the inlet; and an arithmetic unit is placed in the accommodating space and electrically connected to the microphone circuit board, using to receive the acoustic and electrical signals, detect the intensity of the acoustic and electrical signals, and output the original azimuth at the center of symmetry between the azimuth angles of the microphones according to the intensity of the acoustic and electrical signals, wherein the maximum distance between the microphones is greater than is equal to 1.3 times this vertical distance. The sound source localization device according to claim 1, wherein the maximum distance between the microphones is greater than or equal to 2.38 times the vertical distance. The sound source localization device according to claim 1, wherein the maximum distance between the microphones is less than or equal to 3.92 times the vertical distance. A sound source locating device for detecting the original azimuth angle of an original sound source relative to the sound source locating device, which comprises: a top cover with a sliding cavity, the inner side of which is provided with sliding tracks; an entrance A part, placed in the sliding cavity, can be manually or electrically actuated to move on the sliding track, and has one or more through-channels for air circulation, wherein the one or more through-channels have an inlet distributed along the length caliber; a base with an accommodating space and designed to be closely combined with the top cover; a microphone circuit board with a plurality of microphones on it, which can be subjected to changes in air pressure to generate acoustic and electrical signals and communicate with the one or more through-track distances A vertical distance is placed in the accommodating space; and an arithmetic unit is placed in the accommodating space and electrically connected to the microphone circuit board for receiving the acoustic and electrical signals, detecting the intensity of the acoustic and electrical signals, and The original azimuth angle is output at the center of symmetry between the microphone azimuth angles according to the strength of the acoustic and electrical signals, wherein the vertical distance increases or decreases when the entrance part moves on the sliding track, but the entrance part must be fixed on the sliding track. There are at least two places on the track, and the vertical distance corresponding to the two places has a minimum value and a maximum value, and the minimum value is at least three times larger than the entrance aperture, used to detect an original sound source relative to the The original azimuth of the sound source localization device. The sound source positioning device according to claim 4, further comprising: a motor electrically connected to the sliding track and placed in the accommodating space for adjusting the height of the entrance part relative to the sliding track by a control signal, wherein , the operation unit also receives the control signal, and when the intensity distribution of the acoustic and electrical signals has two extreme values, it outputs the result when the inlet component is higher, and vice versa, outputs the result when the inlet component is lower.

Images