Sound source determination device, sound source determination method, and sound source determination program

By using a sound source determination device in the elevator shaft, using a microphone array to collect sound data and combining specific methods in the shaft to calculate the probability density of the sound source direction, the problem of difficulty in determining the sound source in the elevator shaft is solved, and the sound source is realized Accurate identification.

CN115552519BActive Publication Date: 2024-11-05MITSUBISHI ELECTRIC BUILDING SOLUTIONS CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202080100590.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-05-11
Publication Date
2024-11-05
Estimated Expiration
2040-05-11

AI Technical Summary

Technical Problem

In the elevator shaft, it is difficult to determine the location of the sound source because the sound is easily reflected.

Method used

The sound source determination device is adopted, which collects sound data in the elevator shaft through a microphone array, calculates the probability density of the sound source direction in combination with a pre-set method, and uses the position and orientation information of the microphone array, and combines the position information of the car to determine the Sound source location.

Benefits of technology

It effectively solves the problem that the sound source is difficult to determine in the elevator shaft, and realizes the accurate identification and location determination of the sound source.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115552519B_ABST
    Figure CN115552519B_ABST
Patent Text Reader

Abstract

The sound source determination device (100) comprises: an acquisition unit (120) which acquires sound data representing the sound in a shaft in which an elevator car (503) moves, information for determining the position of the car (503), microphone array information (112) representing the position and orientation of a microphone array (200), and sound source determination information (113); a calculation unit (10) which calculates a plurality of probability densities based on the sound data and a predetermined method, and calculates a first inverse, the first inverse being the inverse of a maximum value among the plurality of probability densities or the inverse of a variance value based on the plurality of probability densities; a car position determination unit (160) which determines the position of the car (503) based on the information for determining the position of the car (503); and a sound source determination unit (170) which determines the sound source based on the relationship between the first threshold and the first inverse, the microphone array information (112), the position of the car (503), and the sound source determination information (113).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a sound source determining device, a sound source determining method and a sound source determining program. Background Art

[0002] A microphone is known. Sound is input to the microphone. Here, a technique for estimating the direction of arrival of sound has been proposed (see Patent Document 1). The sound source position estimation device of Patent Document 1 estimates the direction of arrival of sound.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2010-213091 Summary of the invention

[0006] Problems to be solved by the invention

[0007] However, in the elevator shaft, the sound is mostly reflected. In an environment where the sound is mostly reflected, it is difficult to determine the sound source.

[0008] An object of the present invention is to easily determine the source of a sound.

[0009] Means for solving problems

[0010] A sound source determination device according to one embodiment of the present invention is provided. The sound source determination device comprises: an acquisition unit that acquires sound data representing a sound in a shaft where an elevator car moves, information for determining the position of the car, microphone array information representing the position of a microphone array to which the sound is input and the direction of the microphone array, and information for determining the sound source of the sound, i.e., sound source determination information; a calculation unit that calculates a plurality of probability densities representing probability densities of a plurality of sound source direction candidates, respectively, based on the sound data and a predetermined method, and calculates a first inverse, the first inverse being the inverse of a maximum value among the plurality of probability densities or the inverse of a variance value based on the plurality of probability densities; a car position determination unit that determines the position of the car based on the information for determining the position of the car; and a sound source determination unit that determines the sound source based on a relationship between a predetermined first threshold and the first inverse, the microphone array information, the position of the car, and the sound source determination information.

[0011] Effects of the Invention

[0012] According to the present invention, the sound source can be easily determined. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a diagram showing an example of the sound source identification system according to the first embodiment.

[0014] Figure 2 This is a diagram showing a specific example of a case where the microphone array according to Embodiment 1 is installed above a car.

[0015] Figure 3 This is a diagram showing functional blocks included in the sound source identification device according to the first embodiment.

[0016] Figure 4 This is a diagram showing an example of sound source identification information according to the first embodiment.

[0017] Figure 5 This is a flowchart showing the process of recording sound according to the first embodiment.

[0018] Figure 6 This is a flowchart showing an example of processing executed by the sound source identification device according to Embodiment 1.

[0019] Figure 7 This is a diagram showing functional blocks included in the sound source identification device according to the second embodiment.

[0020] Figure 8 This is a diagram showing an example of sound source identification information according to the second embodiment.

[0021] Fig. 9 This is an example of a horizontal cross-sectional view of a hoistway in the second embodiment.

[0022] Fig.10 This is a diagram showing an example of the score in each direction according to the second embodiment.

[0023] Fig.11 (A) and (B) are diagrams showing examples of multiple patterns according to the second embodiment.

[0024] Fig.12 This is a diagram showing functional blocks included in a sound source identification device according to a third embodiment.

[0025] Fig.13 This is a diagram showing an example of sound source identification information according to the third embodiment. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments will be described with reference to the drawings. The following embodiments are merely examples, and various modifications can be made within the scope of the present invention.

[0027] Implementation Method 1

[0028] Figure 1 1 is a diagram showing an example of a sound source identification system according to Embodiment 1. The sound source identification system includes a sound source identification device 100 , a microphone array 200 , an output device 300 , and a driving device 400 .

[0029] The sound source identification device 100 is a device that executes a sound source identification method. For example, the sound source identification device 100 is a device installed in an elevator.

[0030] The microphone array 200 includes a plurality of microphones. In the microphone array 200, the sound data outputted from the plurality of microphones are outputted in a completely synchronized state. In addition, the microphone array 200 may also be equipped with a vibration sensor, an image sensor, an acceleration sensor, an angular acceleration sensor, or a GPS (Global Positioning System) sensor, etc. In addition, these sensors may also be arranged near the microphone array 200.

[0031] For example, the output device 300 is a display. For example, the display is an LCD (Liquid Crystal Display). Alternatively, the output device 300 may be a speaker.

[0032] The driving device 400 is a device for moving the sound source determination device 100. For example, the driving device 400 is a motor and a roller. In addition, when the user installs the sound source determination device 100 in an elevator, the sound source determination system does not include the driving device 400.

[0033] Next, the hardware included in the sound source identifying apparatus 100 will be described.

[0034] The sound source identification device 100 includes a processor 101 , a volatile storage device 102 , a nonvolatile storage device 103 , and a communication device 104 .

[0035] The processor 101 controls the entire sound source determination device 100. For example, the processor 101 is a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field Programmable Gate Array), etc. The processor 101 may also be a multiprocessor. The sound source determination device 100 may also include a processing circuit instead of the processor 101. The processing circuit may also be a single circuit or a composite circuit.

[0036] The volatile storage device 102 is a main storage device of the sound source determination device 100. For example, the volatile storage device 102 is a RAM (Random Access Memory). The nonvolatile storage device 103 is an auxiliary storage device of the sound source determination device 100. For example, the nonvolatile storage device 103 is a ROM (Read Only Memory), a HDD (Hard Disk Drive), or an SSD (Solid State Drive). The volatile storage device 102 and the nonvolatile storage device 103 store various data and programs.

[0037] For example, the communication device 104 is a wired LAN (Local Area Network) adapter, a wireless LAN adapter, or a Bluetooth (registered trademark) adapter. The communication device 104 may also be called a communication interface. The communication device 104 communicates with an external device.

[0038] The microphone array 200 is disposed in the hoistway of the elevator. For example, the microphone array 200 is disposed above the car, below the car, above the counterweight, or below the counterweight. In addition, when the microphone array 200 is disposed above the car, the microphone array 200 is linked to the car. Here, the car may also be referred to as an elevator car. The hoistway may also be referred to as an elevator hoistway.

[0039] In addition, the microphone array 200 may be installed in the pit, near the traction machine, or in a place where the position does not change due to the lifting and lowering operation of the elevator. The microphone array 200 may also be fixed to the hoistway. In addition, the microphone array 200 carried by the user may also be installed in the hoistway by the user. Here, the case where the microphone array 200 is installed above the car is exemplified.

[0040] Figure 2 This is a diagram showing a specific example of a case where the microphone array according to Embodiment 1 is installed above a car. Figure 2 A wall 501 , a car guide rail 502 , a car 503 , an upper car guide shoe 504 , a lower car guide shoe 505 , a counterweight guide rail 506 , a counterweight 507 , an upper counterweight guide shoe 508 , and a lower counterweight guide shoe 509 are shown.

[0041] The wall surface 501 is a concrete wall surface. The car guide rail 502 is a rail for moving the car 503 up and down. The car upper guide shoe 504 and the car lower guide shoe 505 are the joints between the car guide rail 502 and the car 503.

[0042] The counterweight guide rail 506 is a rail for moving the counterweight 507 up and down. The counterweight upper guide shoe 508 and the counterweight lower guide shoe 509 are the joints between the counterweight guide rail 506 and the counterweight 507.

[0043] Next, the functions of the sound source identifying apparatus 100 will be described.

[0044] Figure 3 1 is a diagram showing functional blocks included in the sound source identification device of Embodiment 1. The sound source identification device 100 includes a storage unit 110, an acquisition unit 120, a score calculation unit 130, an accuracy calculation unit 140, an obstacle determination unit 150, a car position determination unit 160, a sound source identification unit 170, and an output unit 180. The acquisition unit 120 and the score calculation unit 130 are included in the calculation unit 10.

[0045] The storage unit 110 may be realized as a storage area reserved in the volatile storage device 102 or the nonvolatile storage device 103 .

[0046] Part or all of the acquisition unit 120, the score calculation unit 130, the accuracy calculation unit 140, the obstacle determination unit 150, the car position determination unit 160, the sound source determination unit 170, and the output unit 180 may also be implemented by the processor 101 or the processing circuit. In addition, part or all of the acquisition unit 120, the score calculation unit 130, the accuracy calculation unit 140, the obstacle determination unit 150, the car position determination unit 160, the sound source determination unit 170, and the output unit 180 may also be implemented as a module of a program executed by the processor 101. For example, the program executed by the processor 101 is also called a sound source determination program. For example, the sound source determination program is recorded in a recording medium.

[0047] The storage unit 110 stores a shaft layout 111 , microphone array information 112 , and sound source identification information 113 .

[0048] The hoistway layout 111 is information indicating the layout of the hoistway. For example, the hoistway layout 111 indicates the wall 501, the car guide rail 502, the car 503, the counterweight guide rail 506, the counterweight 507, the traction machine, the pulley, and the door on the landing side of each floor. For example, this information is expressed in coordinates.

[0049] The microphone array information 112 indicates the position where the microphone array 200 is installed and the direction of the microphone array 200. For example, the microphone array information 112 indicates that the microphone array 200 is installed above the car 503 and the front of the microphone array 200 is in the same direction as the landing side. In addition, the direction of the landing side is the direction in which the door is opened.

[0050] The microphone array information 112 may also be generated by the user operating the sound source determination device 100. The microphone array information 112 may also be generated based on information obtained from an image sensor or a GPS sensor mounted on the microphone array 200. In the case where the microphone array 200 is equipped with a beacon, the microphone array information 112 may also be generated based on information obtained by using the beacon. In addition, the microphone array information 112 may also be generated as follows. The microphone array 200 is equipped with an air pressure sensor. Different air pressure sensors are provided at the bottom of the shaft. The microphone array information 112 is generated based on the comparison result of the values ​​of the two air pressure sensors.

[0051] The sound source identification information 113 is information for identifying the sound source of the sound. Here, the sound source identification information 113 is exemplified.

[0052] Figure 4 1 is a diagram showing an example of sound source identification information in Embodiment 1. Sound source identification information 113 is stored in storage unit 110. Sound source identification information 113 is generated in advance. Sound source identification information 113 includes items of No., relationship, microphone array position, microphone array orientation, car position, obstacle, and sound source. For example, an identifier is registered in the item of No. In addition, for example, information indicating whether an obstacle exists is registered in the item of obstacle.

[0053] The method of using the sound source identification information 113 will be described later. When the hoistway layout 111 is changed, the information registered in the sound source identification information 113 is changed.

[0054] return Figure 3 , the acquisition unit 120 is described.

[0055] The acquisition unit 120 determines the sound data. For example, the acquisition unit 120 acquires the sound data from the microphone array 200. In addition, for example, the acquisition unit 120 acquires the sound data from a recording medium in which the sound data is recorded. When the acquisition unit 120 acquires the sound data from the recording medium, the sound source identification device 100 can be regarded as a PC (Personal Computer).

[0056] The audio data is data representing the sound in the hoistway where the elevator car 503 moves. The audio data is multi-channel audio data.

[0057] The acquisition unit 120 acquires the shaft layout 111, the microphone array information 112, and the sound source determination information 113. For example, the acquisition unit 120 acquires the shaft layout 111, the microphone array information 112, and the sound source determination information 113 from the storage unit 110. Here, the shaft layout 111, the microphone array information 112, and the sound source determination information 113 may also be stored in an external device that can be connected to the sound source determination device 100. For example, the external device is a server. When the shaft layout 111, the microphone array information 112, and the sound source determination information 113 are stored in the external device, the acquisition unit 120 acquires the shaft layout 111, the microphone array information 112, and the sound source determination information 113 from the external device.

[0058] The acquisition unit 120 acquires information for determining the position of the car 503. The information for determining the position of the car 503 is information input by a user to the sound source determination device 100 or information received by the sound source determination device 100 from a device such as the microphone array 200. The information for determining the position of the car 503 will be described later.

[0059] The function of the calculation unit 10 will be described using the score calculation unit 130 and the accuracy calculation unit 140 .

[0060] The score calculation unit 130 calculates a plurality of scores representing the scores of a plurality of sound source direction candidates based on the sound data and a predetermined method. For example, the predetermined method is beam forming, delayed sum method, maximum likelihood method, minimum variance method, MUSIC (MUltiple SIgnal Classification) method, root-MUSIC method, minimum norm method or a learned model. In addition, the predetermined method is the score calculation process shown below. Here, the sound source direction candidate is sometimes expressed only as a direction.

[0061] The calculation process of the score is described in detail. The score calculation unit 130 calculates the score of each horizontal angle with a preset horizontal resolution. For example, when the horizontal resolution is 10 degrees and the object range is 360 degrees, the score calculation unit 130 calculates the scores of 36 directions. In this way, the score calculation unit 130 calculates the scores of each of the multiple directions. Here, the score is expressed as a probability density. Moreover, the score of each direction is expressed as a decimal in such a way that the value obtained by summing up the scores of each direction, that is, the total value is 1.0. In addition, the score represents the probability that a sound source exists in the direction corresponding to the score.

[0062] In addition, the score calculation unit 130 can calculate the signal level in each direction using beam forming, delayed sum method, maximum likelihood method, minimum variance method, MUSIC method, root-MUSIC method, minimum norm method or a learned model. When the score calculation unit 130 calculates the signal level in each direction, it calculates the score in each direction using equation (1) or equation (2). i Indicates the signal level in each direction. i represents the score of each direction. N represents the number of directions. In addition, y i =1(=Σy i ).

[0063]

Mathematical formula 1

[0064]

[0065]

Mathematical formula 2

[0066]

[0067] The accuracy calculation unit 140 calculates the variance value Var of the sound source direction. The variance value Var may also be the maximum value among the scores of each direction. In addition, the accuracy calculation unit 140 may also calculate the variance value Var based on the scores of each direction. In detail, the accuracy calculation unit 140 calculates the y value of the direction with the highest score. i The accuracy calculation unit 140 calculates the variance value Var using the formula (3) when calculating the variance value Var. i represents the horizontal angle. For example, d0 represents 0 degrees. In addition, for example, d 35 represents 350 degrees. Function Diff(d i d j ) represents the horizontal angle d i Angle d with the horizontal j In addition, let y i The direction of maximum value is d MAX .

[0068]

Mathematical formula 3

[0069]

[0070] In addition, for example, the function Diff(d 10 , d 20 ) is 10. Function Diff(d 10 , d 350 ) is 20. The horizontal angle is 0 degree or 360 degrees. Therefore, the accuracy calculation unit 140 calculates the difference in angles in consideration of the surround.

[0071] Next, the accuracy calculation unit 140 calculates the reciprocal of the variance value Var as the accuracy Acc. Specifically, the accuracy calculation unit 140 calculates the accuracy Acc using Equation (4).

[0072]

Mathematical formula 4

[0073] Acc = 1 / Var…(4)

[0074] In this way, the calculated accuracy Acc is the reciprocal of the maximum value among the scores in each direction or the reciprocal of the variance value Var based on the scores in each direction. In addition, the accuracy Acc is also referred to as the first reciprocal.

[0075] In addition, the accuracy Acc can also be expressed as follows. The accuracy Acc is the reciprocal of the standard deviation of the direction of the maximum score among the scores in each direction.

[0076] The obstacle determination unit 150 uses the accuracy Acc and two thresholds to determine whether there is an obstacle between the sound source and the microphone array 200. The two thresholds are Acc1 and Acc2. Acc1 and Acc2 are preset values. Acc1 and Acc2 are stored in the storage unit 110. In addition, Acc2 is larger than Acc1 (i.e., Acc1 < Acc2). In addition, for example, the obstacle is the car body 503 or the counterweight 507.

[0077] In the case of "Acc < Acc1", the obstacle determination unit 150 determines that the state has low accuracy. The state of low accuracy is a state in which the sound emitted from the sound source is reflected and diffused by the wall surface 501 or the like, and the direction of the sound source cannot be determined. The obstacle determination unit 150 determines that the sound source exists farther away than the microphone array 200. This is because the sound emitted from a distance is reflected by the wall surface 501, and this sound reaches all directions of the microphone array 200 with substantially the same signal level, and this sound reaches the microphone array 200 with a random phase difference.

[0078] In the case of "Acc < Acc1", the obstacle determination unit 150 determines that there is no obstacle between the sound source and the microphone array 200. In addition, in the case of "Acc < Acc1", the obstacle determination unit 150 can also be determined to have an obstacle through the user's setting.

[0079] When "Acc1 ≤ Acc < Acc2", the obstacle determination unit 150 determines that the state is accurate to a certain extent. The state that is accurate to a certain extent is the state in which the sound emitted from the sound source is restricted by an obstacle to a certain extent and input to the microphone array 200. For example, the obstacle determination unit 150 obtains the position of the car 503 and the position of the counterweight 507 determined by the car position determination unit 160 described later, the hoistway layout 111, and the microphone array information 112. When the position of the car 503 is at the upper part of the hoistway, the obstacle determination unit 150 determines that the car 503 exists as an obstacle between the sound source and the microphone array 200.

[0080] When "Acc1 ≤ Acc < Acc2", the obstacle determination unit 150 determines that there is an obstacle between the sound source and the microphone array 200.

[0081] When "Acc2 ≤ Acc", the obstacle determination unit 150 determines that the state is highly accurate. The highly accurate state is the state in which the distance between the sound source and the microphone array 200 is short. When "Acc2 ≤ Acc", the obstacle determination unit 150 determines that there is no obstacle between the sound source and the microphone array 200.

[0082] The car position determination unit 160 determines the vertical position of the car 503 based on the information for determining the position of the car 503. Here, for example, the information for determining the position of the car 503 represents the speed of the car 503 and the moving time of the car 503. Here, for example, the initial position of the car 503 is preset in advance. The car position determination unit 160 determines the position of the car 503 based on the moving time of the car 503 from the initial position (i.e., the moving time of the car 503) and the speed of the car 503. In addition, for example, the information for determining the position of the car 503 represents the information obtained from the acceleration sensor and the moving time of the car 503. The car position determination unit 160 can also determine the position of the car 503 based on the information obtained from the acceleration sensor and this time. In addition, for example, the information for determining the position of the car 503 is the radio wave intensity of the WiFi (registered trademark) access point, the radio wave intensity of the Bluetooth beacon, or the information obtained from the image sensor. The car position determination unit 160 can also determine the position of the car 503 based on the radio wave intensity of the WiFi access point, the radio wave intensity of the Bluetooth beacon, or the information obtained from the image sensor. The car position determination unit 160 can also use existing technologies to determine the position of the car 503.

[0083] In addition, the car position determination unit 160 determines the position of the counterweight 507. Here, the car 503 and the counterweight 507 are connected by ropes. Therefore, when the position of the car 503 is determined, the approximate position of the counterweight 507 is determined. However, the positional relationship between the car 503 and the counterweight 507 varies depending on the individual car. Therefore, the shaft layout 111 is used. Thus, when the car position determination unit 160 determines the position of the counterweight 507, it determines the position of the counterweight 507 based on the position of the car 503 and the shaft layout 111.

[0084] The sound source identification unit 170 identifies the sound source using the sound source identification information 113. Here, the method of using the sound source identification information 113 is described. The sound source identification unit 170 retrieves the record that matches the relationship between the accuracy Acc calculated by the accuracy calculation unit 140 and the threshold Acc1 and the threshold Acc2, the position and orientation of the microphone array 200 shown in the microphone array information 112, the position of the car 503 determined by the car position identification unit 160, and the determination result of the obstacle determination unit 150 from the sound source identification information 113. The information registered in the sound source item of the retrieved record is the sound source. For example, when the retrieved record is the record No. 1, the sound source identification unit 170 identifies the device that exists above the car 503 as the sound source. In addition, for example, the device is the counterweight 507, the guide shoe, the traction machine, etc. In this way, the sound source identification information 113 indicates the relationship between the accuracy Acc and the threshold Acc1 and the threshold Acc2, the position and orientation of the microphone array 200, the position of the car 503, and the information indicating whether there is an obstacle. Furthermore, the sound source identification unit 170 can identify the sound source using the sound source identification information 113 .

[0085] Here, the sound source identification information 113 may not include an obstacle item. When the sound source identification information 113 does not include an obstacle item, the sound source identification unit 170 retrieves from the sound source identification information 113 a record that is consistent with the relationship between the accuracy Acc calculated by the accuracy calculation unit 140 and the threshold values ​​Acc1 and Acc2, the position and orientation of the microphone array 200 shown in the microphone array information 112, and the position of the car 503 determined by the car position determination unit 160.

[0086] In addition, the threshold value Acc1 or the threshold value Acc2 indicated in the items No. 1, No. 4, and No. 5 of the sound source identification information 113 may also be referred to as the first threshold value. The threshold value Acc1 indicated in the items No. 2 and No. 3 of the sound source identification information 113 may also be referred to as the first threshold value. The threshold value Acc2 indicated in the items No. 2 and No. 3 of the sound source identification information 113 may also be referred to as the second threshold value. The threshold value Acc2 indicated in the items No. 4 and No. 5 of the sound source identification information 113 may also be referred to as the second threshold value.

[0087] Furthermore, the sound source identification information 113 may indicate the relationship between the accuracy Acc and the threshold Acc1 and the relationship between the accuracy Acc and the threshold Acc2. That is, the sound source identification information 113 may not include the records of No. 2 and No. 3.

[0088] In addition, when the retrieved record is a record of No. 4 or No. 5, the sound source determination unit 170 determines that the device existing in the sound source direction is the sound source. In this case, the sound source determination unit 170 sets the direction corresponding to the maximum value of the scores of each direction calculated by the score calculation unit 130 as the sound source direction. The sound source determination unit 170 determines the sound source from the shaft layout 111 based on the position of the microphone array 200 shown in the microphone array information 112, the sound source direction, and the shaft layout 111.

[0089] The output unit 180 outputs the information indicating the sound source to the output device 300. For example, when the output device 300 is a display, the output unit 180 outputs the information indicating the sound source to the display. Thus, the display displays the information indicating the sound source. Furthermore, for example, when the output device 300 is a speaker, the output unit 180 outputs the information indicating the sound source to the speaker. Thus, the speaker outputs the information indicating the sound source using voice.

[0090] Next, a series of processes will be described using a flowchart. First, for example, the user sets the microphone array 200 in the shaft. The user inputs the position where the microphone array 200 is set and the direction of the microphone array 200 to the sound source determination device 100. As a result, the sound source determination device 100 generates microphone array information 112. Then, the sound collection preparation is completed. The sound collection is described using a flowchart.

[0091] Figure 5 This is a flowchart showing the sound recording process of embodiment 1.

[0092] (Step S11 ) The microphone array 200 starts recording.

[0093] (Step S12) The car 503 moves in the hoistway. Specifically, the car 503 ascends or descends.

[0094] The sound in the hoistway is input to the microphone array 200 .

[0095] (Step S13) The microphone array 200 ends recording.

[0096] Next, the process executed by the sound source identifying apparatus 100 will be described using a flowchart.

[0097] Figure 6This is a flowchart showing an example of processing executed by the sound source identification device according to Embodiment 1.

[0098] (Step S21 ) The acquisition unit 120 acquires audio data and the like of audio recorded by the microphone array 200 .

[0099] (Step S22 ) The score calculation unit 130 calculates the scores of each of the plurality of directions using the sound source data.

[0100] (Step S23 ) The accuracy calculation unit 140 calculates the variance value Var using the scores of each of the plurality of sound source direction candidates.

[0101] (Step S24 ) The accuracy calculation unit 140 calculates the accuracy Acc using the variance value Var.

[0102] (Step S25) The car position specifying unit 160 specifies the position of the car 503 using the information for specifying the position of the car 503.

[0103] (Step S26 ) The obstacle determination unit 150 determines the presence of an obstacle using the accuracy Acc, the threshold value Acc1 , and the threshold value Acc2 .

[0104] (Step S27 ) The sound source identification unit 170 identifies the sound source using the sound source identification information 113 .

[0105] (Step S28 ) The output unit 180 outputs information indicating the sound source to the output device 300 .

[0106] According to Embodiment 1, the sound source identifying device 100 can identify the sound source using the sound data representing the sound in the hoistway. Furthermore, the sound source identifying device 100 identifies the sound source using the sound source identifying information 113 without performing complicated processing. Thus, the sound source identifying device 100 can easily identify the sound source.

[0107] Implementation Method 2

[0108] Next, Embodiment 2 will be described. In Embodiment 2, matters different from Embodiment 1 will be mainly described. In addition, in Embodiment 2, the description of matters common to Embodiment 1 will be omitted. In the description of Embodiment 2, reference will be made to Figures 1 to 3 .

[0109] In the first embodiment, two threshold values ​​(threshold value Acc1 and threshold value Acc2) are used to determine the presence of an obstacle. In the second embodiment, a case where one threshold value (threshold value Acc2) is used to determine the presence of an obstacle will be described.

[0110] Figure 7 FIG. 2 is a diagram showing functional blocks of a sound source identification device according to Embodiment 2. Figure 3 The same structure as shown Figure 7 Structural annotation and Figure 3 The same reference numerals are shown.

[0111] The sound source identification device 100a includes a storage unit 110a, an acquisition unit 120a, an accuracy calculation unit 140a, an obstacle determination unit 150a, a sound source identification unit 170a, and a pattern identification unit 191. The score calculation unit 130 and the accuracy calculation unit 140a are included in the calculation unit 10a.

[0112] The storage unit 110a stores sound source identification information 113a. Here, the sound source identification information 113a is exemplified.

[0113] Figure 8 11 is a diagram showing an example of sound source identification information according to Embodiment 2. The sound source identification information 113a is information in which a pattern item is added to the sound source identification information 113. The pattern will be described later.

[0114] The information registered in the item related to the sound source identification information 113a is different from the information registered in the item related to the sound source identification information 113. Here, the threshold value Acc2 registered in the item related to the sound source identification information 113a is also referred to as a first threshold value.

[0115] return Figure 7 , the sound source determination device 100a is described.

[0116] The score calculation unit 130 calculates the score of each direction. The accuracy calculation unit 140a determines whether there is a direction in a predetermined interval with a score higher than the average of the scores of each direction among the plurality of directions. Fig. 9 and Fig.10 This determination will be described.

[0117] Fig. 9 This is an example of a horizontal cross-sectional view of a hoistway in the second embodiment. Fig. 9 The microphone array 200, the car guide rail 502, the counterweight guide rail 506 and the counterweight 507 are shown. Fig. 9 The front direction of the microphone array 200 is shown.

[0118] Fig.10 This is a diagram showing an example of the score in each direction according to the second embodiment. Fig.10 Shown according to the expression in Fig. 9 The scores of each direction are calculated from the sound data of the sound recorded in the state of . The solid line shows the score of each direction. The dotted line shows the average of the scores of each direction.

[0119] For example, in Fig.10Among them, the direction in which the score higher than the average of the scores in each direction exists within the specified range is the back surface. When there is a direction in which the score higher than the average of the scores in each direction exists within the specified range among multiple directions, the accuracy calculation unit 140a sends the scores in each direction to the obstacle determination unit 150a.

[0120] When "Acc < Acc2", the obstacle determination unit 150a determines that there is an obstacle between the sound source and the microphone array 200. In addition, the pattern determination unit 191 compares the scores in each direction calculated by the score calculation unit 130 with multiple patterns of multiple scores. These multiple scores are also referred to as multiple first probability densities. Here, multiple patterns are illustrated.

[0121] Fig.11 (A) and (B) of FIG. are diagrams showing examples of multiple patterns of Embodiment 2. Fig.11 Let (A) of FIG. be Pattern 1. Fig.11 Let (B) of FIG. be Pattern 2. In Fig.11 FIG., two patterns are shown. The number of patterns may also be three or more. Additionally, for example, multiple patterns are stored in the storage unit 110.

[0122] The acquisition unit 120a acquires information indicating multiple patterns.

[0123] The pattern determination unit 191 determines a pattern suitable for the scores in each direction calculated by the score calculation unit 130 from among the multiple patterns. For example, the pattern determination unit 191 determines which of Pattern 1 and Pattern 2 the scores in each direction calculated by the score calculation unit 130 are suitable for. In addition, the pattern determination unit 191 may also use the mean square error to determine which pattern the scores in each direction calculated by the score calculation unit 130 are suitable for.

[0124] When "Acc2 ≤ Acc", the obstacle determination unit 150a determines that there is no obstacle between the sound source and the microphone array 200. In addition, when "Acc2 ≤ Acc", the pattern determination unit 191 does not perform the above-described suitability process for the multiple patterns.

[0125] The sound source determination unit 170a retrieves a record from the sound source determination information 113a that is consistent with the relationship between the accuracy Acc calculated by the accuracy calculation unit 140 and the threshold Acc2, the pattern determined by the pattern determination unit 191, the position and orientation of the microphone array 200 shown in the microphone array information 112, the position of the car 503 determined by the car position determination unit 160, and the determination result of the obstacle by the obstacle determination unit 150a. The information registered in the sound source item of the retrieved record is the sound source. In this way, the sound source determination unit 170a can determine the sound source using the sound source determination information 113a.

[0126] In the second embodiment, Acc1 is not used. Therefore, the user does not need to perform the operation of setting Acc1. Therefore, according to the second embodiment, the burden on the user is reduced.

[0127] Implementation 3

[0128] Next, Embodiment 3 will be described. In Embodiment 3, matters different from Embodiment 1 will be mainly described. Also, in Embodiment 3, matters similar to Embodiment 1 will be omitted. In the description of Embodiment 3, reference will be made to Figures 1 to 6 .

[0129] In Embodiments 1 and 2, a case where one microphone array is used is described. In Embodiment 3, a case where a plurality of microphone arrays are used is described.

[0130] Fig.12 1 is a diagram showing functional blocks of a sound source identification device according to Embodiment 3. The sound source identification device 100b is connected to microphone arrays 200 and 201. For example, the microphone array 200 is installed above the car 503. The microphone array 201 is installed above the counterweight 507. Fig.12 In FIG. 2 , two microphone arrays are illustrated. The number of microphone arrays may be three or more. In addition, when recording sound, a synchronization signal is input to the microphone arrays 200 and 201. That is, in step S11, a synchronization signal is input to the microphone arrays 200 and 201.

[0131] The sound source identification device 100b includes a storage unit 110b, an acquisition unit 120b, and a synchronization processing unit 192. Figure 3 The same structure as shown Fig.12 Structural annotation and Figure 3 The storage unit 110b stores microphone array information 112b and sound source identification information 113b.

[0132] The microphone array information 112b indicates the positions of the microphone arrays 200 and 201 and the directions of the microphone arrays 200 and 201. Next, the sound source identification information 113b is exemplified.

[0133] Fig.13 11 is a diagram showing an example of sound source identification information according to Embodiment 3. In the sound source identification information 113b, items such as the relationship between the number of microphone arrays, the position of the microphone array, the direction of the microphone array, the position of the car, and obstacles are added.

[0134] Fig.13 Items such as the relationship corresponding to the microphone array 1 are shown. In addition, the microphone array 1 refers to the microphone array 200. In addition, Fig.13Items such as the relationship corresponding to the microphone array 2 are shown. In addition, the microphone array 2 refers to the microphone array 201.

[0135] The method of using the sound source identification information 113b is briefly described. The sound source identification device 100b retrieves the corresponding row from the column of the microphone array 1 based on the sound data representing the sound input to the microphone array 200. The sound source identification device 100b retrieves the corresponding row from the column of the microphone array 2 based on the sound data representing the sound input to the microphone array 201. The information registered in the item of the sound source recorded in the two rows that match is the sound source. In this way, the sound source identification device 100b identifies the sound source.

[0136] The acquisition unit 120 b acquires audio data representing audio input to the microphone array 200 and audio data representing audio input to the microphone array 201 .

[0137] Furthermore, the acquisition unit 120b acquires information for specifying the position of the car 503, the microphone array information 112b, and the sound source specifying information 113b.

[0138] The synchronization processing unit 192 uses the synchronization signal as a reference to synchronize the audio data representing the audio input to the microphone array 200 and the audio data representing the audio input to the microphone array 201 .

[0139] The calculation unit 10, the obstacle determination unit 150, and the car position determination unit 160 perform the same processing as in Embodiment 1 for each of the plurality of sound data. For example, the calculation unit 10 calculates the accuracy Acc corresponding to each of the two sound data based on the sound data representing the sound input to the microphone array 200 and the sound data representing the sound input to the microphone array 201.

[0140] The sound source determination unit 170b determines the sound source based on the relationship between the accuracy Acc and the threshold Acc1 and the threshold Acc2 respectively corresponding to the two sound data, the microphone array information 112b, the position of the car 503 determined by the car position determination unit 160, the determination result of the obstacle determination unit 150 and the sound source determination information 113b.

[0141] Furthermore, the sound source identification information 113 b may not include an item of an obstacle. When the sound source identification information 113 b does not include an item of an obstacle, the sound source identification unit 170 b identifies the sound source without using the determination result of the obstacle determination unit 150 .

[0142] According to Embodiment 3, the sound source identification apparatus 100b identifies the sound source using the sound data output from the plurality of microphone arrays, and therefore can identify the sound source more accurately.

[0143] The features in the above-described embodiments can be combined with each other as appropriate.

[0144] Description of symbols

[0145] 10, 10a: calculation unit; 100, 100a, 100b: sound source determination device; 101: processor; 102: volatile storage device; 103: non-volatile storage device; 104: communication device; 110, 110a, 110b: storage unit; 111: shaft layout; 112, 112b: microphone array information; 113, 113a, 113b: sound source determination information; 120, 120a, 120b: acquisition unit; 130: score calculation unit; 140, 140a: accuracy calculation unit; 150, 150a: obstacle determination unit; 160: car position determination unit; 170, 170a, 170b: sound source determination unit; 180: output unit; 191: pattern determination unit; 192: synchronization processing unit; 200, 201: microphone array; 300: output device; 400: driving device; 501: wall; 502: car guide rail; 503: car; 504: upper guide shoe; 505: lower guide shoe; 506: counterweight guide rail; 507: counterweight; 508: counterweight upper guide shoe; 509: counterweight lower guide shoe.

Claims

1. A sound source determination device, the sound source determination device comprising: an acquisition unit that acquires sound data representing a sound in a shaft in which an elevator car moves, information for determining a position of the car, microphone array information representing a position of a microphone array to which the sound is input and a direction of the microphone array, and sound source determination information that is information for determining a sound source of the sound; a calculation unit that calculates a plurality of probability densities that respectively represent probability densities of a plurality of sound source direction candidates based on the sound data and a predetermined method, and calculates a first inverse, wherein the first inverse is the inverse of a maximum value among the plurality of probability densities or the inverse of a variance value based on the plurality of probability densities; a car position determining unit that determines the position of the car based on information for determining the position of the car; and A sound source identifying unit identifies the sound source by searching the sound source identifying information based on the relationship between a preset first threshold value and the first inverse, the microphone array information, and the position of the car.

2. The sound source determination device according to claim 1, wherein: The sound source identification device further includes an obstacle determination unit, which determines whether there is an obstacle between the sound source and the microphone array based on the relationship between the first threshold and the first inverse number. The sound source identification unit identifies the sound source based on the relationship between the first threshold value and the first inverse, the microphone array information, the position of the car, the result of the determination, and the sound source identification information.

3. The sound source determination device according to claim 1, wherein: The sound source determination device further comprises a pattern determination unit, The acquisition unit acquires information of a plurality of patterns representing a plurality of first probability densities, When there is a sound source direction candidate in a specified interval having a probability density higher than the average of the plurality of probability densities among the plurality of sound source direction candidates, and the first inverse is smaller than the first threshold, the pattern determination unit determines a pattern suitable for the plurality of probability densities from among the plurality of patterns, The sound source identification unit identifies the sound source based on the relationship between the first threshold value and the first inverse, the identified pattern, the microphone array information, the position of the car, and the sound source identification information.

4. The sound source determination device according to claim 2, wherein: The sound source determination device further comprises a pattern determination unit, The acquisition unit acquires information of a plurality of patterns representing a plurality of first probability densities, When there is a sound source direction candidate in a specified interval having a probability density higher than the average of the plurality of probability densities among the plurality of sound source direction candidates, and the first inverse is smaller than the first threshold, the pattern determination unit determines a pattern suitable for the plurality of probability densities from among the plurality of patterns, The sound source identification unit identifies the sound source based on the relationship between the first threshold value and the first inverse, the identified pattern, the microphone array information, the position of the car, and the sound source identification information.

5. The sound source determination device according to claim 1, wherein: The sound source identification unit identifies the sound source based on the relationship between the first threshold value and a preset second threshold value and the first inverse, the microphone array information, the position of the car, and the sound source identification information. The second threshold is larger than the first threshold.

6. The sound source determination device according to claim 5, wherein: The sound source identification device further includes an obstacle determination unit, which determines whether there is an obstacle between the sound source and the microphone array based on the relationship between the first threshold value, the second threshold value and the first inverse number. The sound source identification unit identifies the sound source based on the relationship between the first threshold value, the second threshold value and the first inverse, the microphone array information, the position of the car, the result of the determination, and the sound source identification information.

7. The sound source determination device according to claim 1, wherein: The sound source determination device further comprises a synchronization processing unit. The acquisition unit acquires a plurality of sound data output from a plurality of microphone arrays to which the sound is input, information for determining the position of the car, the microphone array information indicating the positions of the plurality of microphone arrays and the directions of the plurality of microphone arrays, and the sound source determination information, The synchronization processing unit synchronizes the plurality of audio data. The calculation unit calculates the first inverse number corresponding to each of the plurality of sound data based on the plurality of sound data, The sound source identification unit identifies the sound source based on the relationship between the first inverse corresponding to each of the plurality of sound data and the first threshold, the microphone array information, the position of the car, and the sound source identification information.

8. The sound source determination device according to claim 7, wherein: The sound source determination unit determines the sound source based on the relationship between the first inverse corresponding to each of the plurality of sound data and the first threshold and a preset second threshold, the microphone array information, the position of the car, and the sound source determination information. The second threshold is larger than the first threshold.

9. The sound source identification device according to any one of claims 1 to 8, wherein: The information for determining the position of the car indicates the speed of the car and the moving time of the car. The car position determination unit determines the position of the car based on the speed of the car and the travel time of the car.

10. The sound source identification device according to any one of claims 1 to 8, wherein: The sound source identification device further includes an output unit that outputs information indicating the sound source.

11. The sound source determination device according to claim 9, wherein: The sound source identification device further includes an output unit that outputs information indicating the sound source.

12. A method for determining a sound source, wherein: The sound source identification device obtains sound data representing the sound in the shaft where the elevator car moves, information for identifying the position of the car, microphone array information representing the position of the microphone array into which the sound is input and the orientation of the microphone array, and sound source identification information, which is information for identifying the sound source of the sound. The sound source determination device calculates a plurality of probability densities respectively representing the probability densities of a plurality of sound source direction candidates based on the sound data and a predetermined method, The sound source determination device calculates a first inverse, which is the inverse of a maximum value among the plurality of probability densities or the inverse of a variance value based on the plurality of probability densities. The sound source determining device determines the position of the car based on the information used to determine the position of the car, The sound source identification device identifies the sound source by searching the sound source identification information based on the relationship between the preset first threshold value and the first inverse, the microphone array information, and the position of the car.

13. A computer program product, comprising a sound source determination program, wherein when the sound source determination program is executed by a processor, the sound source determination program performs the following processing: Acquiring sound data representing the sound in a shaft in which an elevator car moves, information for determining the position of the car, microphone array information representing the position of a microphone array into which the sound is input and the orientation of the microphone array, and sound source determination information, which is information for determining the sound source of the sound, Calculating a plurality of probability densities respectively representing the probability densities of a plurality of sound source direction candidates based on the sound data and a predetermined method, Calculating a first inverse, the first inverse is the inverse of the maximum value among the multiple probability densities or the inverse of the variance value based on the multiple probability densities, determining the position of the car based on information for determining the position of the car, The sound source is identified by searching the sound source identification information based on the relationship between the preset first threshold value and the first inverse, the microphone array information, and the position of the car.