Abnormal sound detection system, abnormal sound detection method, and program

Abstract

To provide an abnormal sound detection system, an abnormal sound detection method, and a program capable of more accurately grasping generation of abnormal sound in real time.SOLUTION: An abnormal sound detection system includes: a sound data acquisition unit configured to acquire a sound collected by a sound collection device provided in a target area in which abnormal sound is generated as sound data; an abnormal sound detection unit configured to detect, every time predetermined time elapses, the abnormal sound generated in the target area in the predetermined time based on the sound data of the sound collected in the predetermined time; and a notification control unit configured to gives a notice that the abnormal sound is detected in the target area when detection results of the abnormal sound in the predetermined time satisfies predetermined conditions.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an abnormal sound detection system, an abnormal sound detection method, and a program. [Background technology]

[0002] Conventionally, various techniques have been proposed for detecting a specific sound based on collected sounds.

[0003] For example, Patent Document 1 below discloses a technology capable of detecting vehicle noise. In this technology, the noise generated by each vehicle traveling on a road is collected, and the collected sounds are classified into predetermined noise levels based on the volume of the collected sounds, thereby detecting the noise of each vehicle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2002-92783 A Summary of the Invention [Problem to be solved by the invention]

[0005] In order to respond appropriately to detected abnormal sounds, it is necessary to be able to grasp the occurrence of abnormal sounds more accurately and in real time.

[0006] In view of the above-mentioned problems, an object of the present invention is to provide an abnormal sound detection system, an abnormal sound detection method, and a program that are capable of detecting the occurrence of abnormal sounds more accurately and in real time. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems, an abnormal sound detection system according to one embodiment of the present invention is an abnormal sound detection system that includes: a sound data acquisition unit that acquires, as sound data, sound collected by a sound collection device provided in a target area where abnormal sound occurs; an abnormal sound detection unit that detects, each time a predetermined time elapses, an abnormal sound generated in the target area at a predetermined time based on the sound data of the sound collected at the predetermined time; and a notification control unit that notifies that the abnormal sound has been detected in the target area if the detection result of the abnormal sound at the predetermined time satisfies a predetermined condition.

[0008] An abnormal sound detection method according to one aspect of the present invention is an abnormal sound detection method executed by a computer, the abnormal sound detection method including: a sound data acquisition process for acquiring, as sound data, sound collected by a sound collection device provided in a target area where an abnormal sound occurs; an abnormal sound detection process for detecting, each time a predetermined time elapses, an abnormal sound generated in the target area at a predetermined time based on the sound data of the sound collected at the predetermined time; and a notification control process for notifying that the abnormal sound has been detected in the target area if the detection result of the abnormal sound at the predetermined time satisfies a predetermined condition.

[0009] A program according to one aspect of the present invention is a program for causing a computer to function as: sound data acquisition means for acquiring, as sound data, sound collected by a sound collecting device installed in a target area in which an abnormal sound occurs; abnormal sound detection means for detecting, each time a predetermined time elapses, an abnormal sound generated in the target area at a predetermined time based on the sound data of the sound collected at the predetermined time; and notification control means for notifying that the abnormal sound has been detected in the target area if the detection result of the abnormal sound at the predetermined time satisfies predetermined conditions. Effect of the Invention

[0010] According to the present invention, the occurrence of abnormal sound can be detected more accurately and in real time. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an abnormal sound detection system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram showing an example of a functional configuration of the abnormal sound detection device according to the present embodiment. [Diagram 3] FIG. 4 is a diagram showing an example of a spectrogram image according to the present embodiment. [Figure 4] FIG. 2 is a diagram illustrating an example of a dataset used for learning a detection model according to the present embodiment. [Diagram 5] 2 is a block diagram showing an example of a functional configuration of a management server according to the present embodiment. FIG. [Figure 6] 2 is a block diagram showing an example of a functional configuration of a receiving terminal according to the present embodiment. FIG. [Figure 7] 11 is a diagram showing an example of aggregated data displayed on a receiving terminal according to the embodiment; FIG. [Figure 8] FIG. 2 is a sequence diagram showing an example of a process flow relating to abnormal sound detection processing and notification processing in the abnormal sound detection system according to the embodiment. [Figure 9] FIG. 11 is a sequence diagram showing an example of a process flow related to a tallying process in the abnormal sound detection system according to the embodiment. [Figure 10] FIG. 11 is a sequence diagram showing an example of a process flow related to a prediction process in the abnormal sound detection system according to the embodiment. [Figure 11] FIG. 11 is a sequence diagram showing an example of a processing flow related to weather processing in the abnormal sound detection system according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0013] <1. Configuration of abnormal sound detection system> The configuration of an abnormal sound detection system according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of the configuration of an abnormal sound detection system according to this embodiment.

[0014] The abnormal sound detection system 1 shown in FIG. 1 is a system for detecting an abnormal sound occurring in a target area and for notifying that an abnormal sound has been detected in the target area. The target area is an area where abnormal sounds occur. It may be either outdoors or indoors, and its size is not particularly limited. As an example, the target area according to this embodiment is an area where drifting by vehicles is performed, and drifting sounds due to drifting are generated as noise (hereinafter also referred to as "drift noise"). The vehicle is, for example, an automobile. The abnormal sound is a sound that occurs when an abnormal state occurs in the target area. As an example, the abnormal sound according to the present embodiment is a drift noise. That is, in the target area, a state in which a car is drifting is an abnormal state, and the drift noise that occurs in that state is the abnormal sound. It should be noted that sounds that occur when no abnormal conditions are occurring in the target area are normal sounds. The normal sounds according to this embodiment are, for example, sounds that occur when no drifting is being performed by a car in the target area, i.e., sounds that occur when no drift noise is occurring.

[0015] Drifting of automobiles is likely to take place in areas with characteristics such as many straight roads, wide roads, and few traffic lights (such as industrial districts). In areas where drifting is practiced, drift noise causes noise pollution for nearby residents and anxiety for factory employees who work at night. When drift noise is confirmed, the person who confirms it reports it to the police by phone or other means. However, it is difficult for a person to accurately confirm the occurrence of drift noise, and when there is a time lag between the occurrence of drift noise and the report, the police are unable to crack down adequately.

[0016] Therefore, in this embodiment, the abnormal sound detection system 1 detects drift noise occurring in a target area and notifies the user that drift noise has been detected in the target area. The notification destination can be set arbitrarily, and may be, for example, the police or a local government, or a user such as a nearby resident. In this way, the occurrence of drift noise is detected by the system, not a person, and the detection is notified in real time. Therefore, the police, local government, nearby residents, etc. can grasp the occurrence of drift noise more accurately and in real time. Furthermore, the police can carry out sufficient enforcement against drift noise.

[0017] As shown in FIG. 1, the abnormal sound detection system 1 includes an abnormal sound detection device 10, a management server 20, and a receiving terminal 30. The network NW may be configured to transmit and receive information using, for example, a Local Area Network (LAN), a Wide Area Network (WAN), a telephone network (mobile phone network, landline telephone network, etc.), a regional Internet Protocol (IP) network, the Internet, etc.

[0018] (1) Abnormal sound detection device 10 The abnormal sound detection device 10 is a device that detects drift noise occurring in a target area and notifies that the drift noise has been detected. An abnormal sound detection device 10 is provided for each target area and detects drift noise occurring in each target area. There is no particular limit to the number of abnormal sound detection devices 10 provided in one target area, and one or more abnormal sound detection devices 10 may be provided.

[0019] The abnormal sound detection device 10 is communicatively connected to the management server 20 via the network NW. When the abnormal sound detection device 10 detects the occurrence of drift noise, it transmits a notification to the management server 20 that drift noise has occurred in the target area. When notifying the management server 20, the abnormal sound detection device 10 transmits notification data indicating data related to the detected drift noise to the management server 20. The notification data is data indicating, for example, the ID of the abnormal sound detection device 10, the location (target area) where the drift noise occurred, the date and time of occurrence, etc.

[0020] The abnormal sound detection device 10 detects the occurrence of drift noise using a detection model. If multiple detection models are prepared, the abnormal sound detection device 10 selects an appropriate detection model depending on the situation at the time of detection and detects the occurrence of drift noise. The detection model used by the abnormal sound detection device 10 is determined by, for example, the management server 20. In this case, the abnormal sound detection device 10 receives a switching instruction from the management server 20, and selects and uses the detection model indicated by the switching instruction.

[0021] 1, the abnormal sound detection device 10 includes a sound collection device 11, a power supply device 12, and a control device 13. The power supply device 12 and the control device 13 are stored inside the housing of the abnormal sound detection device 10. The sound collection device 11 is provided outside the housing of the abnormal sound detection device 10, in a state connected to the control device 13 inside the housing of the abnormal sound detection device 10. The housing of the abnormal sound detection device 10 is waterproof.

[0022] (1-1) Sound collection device 11 The sound collector 11 is a device that collects sounds generated in a target area. The sound collector 11 includes, for example, a main microphone and a sub-microphone. The main microphone is installed facing the source of drift noise in order to collect drift noise in the target area. The source of drift noise is a car, but the car is not always present in the target area. For this reason, the main microphone is installed facing a position where drift driving is expected to take place based on tire marks left on the road by drift driving. The sub-microphone is installed facing a different direction from the main microphone in order to collect environmental sounds in the target area. The sound collection device 11 is connected to the control device 13 via a wired connection or a wireless connection so as to be able to communicate with the control device 13. When the sound collection device 11 collects sound in a target area, it converts the collected sound into data and transmits the converted sound data to the control device 13.

[0023] (1-2) Power supply device 12 The power supply device 12 is a power supply source for the abnormal sound detection device 10. The power supply device 12 is, for example, a mobile battery. The power supply device 12 is connected to the control device 13 by a wired connection, and supplies power to the control device 13.

[0024] (1-3) Control device 13 The control device 13 is a device that controls the overall operation of the abnormal sound detection device 10. The control device 13 includes, for example, a control board, a storage medium, a communication device, and the like. The control board performs processing for detecting drift noise occurring in the target area based on sound data of the sound collected by the sound collection device 11 (hereinafter also referred to as "abnormal sound detection processing"). The storage medium stores various data related to the detection of drift noise. The communication device receives a switching instruction from the management server 20 via the network NW, and transmits a notification that drift noise has occurred in the target area to the management server 20.

[0025] (2) Management Server 20 The management server 20 is a server that executes processes for managing the occurrence of drift noise. The management server 20 is configured with one or more servers (e.g., cloud servers). The management server 20 is communicatively connected to the abnormal sound detection device 10 and the receiving terminal 30 via the network NW. In communication with the abnormal sound detection device 10, the management server 20 receives a notification (including notification data) of the occurrence of drift noise, and transmits an instruction to switch the detection model. The management server 20 performs processing based on the notification data received from the abnormal sound detection device 10. In communication with the receiving terminal 30, the management server 20 transmits the notification (including notification data) of the occurrence of drift noise, data generated by processing based on the notification data, and the like.

[0026] (3) Receiving terminal 30 The receiving terminal 30 is a terminal that receives a notification that drift noise has occurred in a target area from the management server 20. The receiving terminal 30 is, for example, a PC (Personal Computer), a smartphone, a tablet terminal, etc. The receiving terminal 30 is connected to the management server 20 via the network NW so as to be able to communicate with the management server 20. There is no particular limit to the number of receiving terminals 30 included in the abnormal sound detection system 1 and their users. The users may be, for example, the police, local governments, local residents, system administrators, and the like.

[0027] <2. Functional configuration of abnormal sound detection device> The configuration of the abnormal sound detection system 1 according to this embodiment has been described above. Next, the functional configuration of the abnormal sound detection device 10 according to this embodiment will be described with reference to Fig. 2 to Fig. 4. Fig. 2 is a block diagram showing an example of the functional configuration of the abnormal sound detection device 10 according to this embodiment. As shown in FIG. 2, the abnormal sound detection device 10 includes a sound collection unit 110, a power supply unit 120, a communication unit 130, a storage unit 140, and a control unit 150.

[0028] (1) Sound collection section 110 The sound collection unit 110 has a function of collecting sounds generated in a target area and converting the collected sounds into sound data. The function of the sound collection unit 110 is realized by the sound collection device 11 provided as hardware in the abnormal sound detection device 10 described with reference to FIG.

[0029] (2) Power supply section 120 The power supply unit 120 has a function of supplying power to the abnormal sound detection device 10. The function of the power supply unit 120 is realized by the power supply device 12 provided as hardware in the abnormal sound detection device 10 described with reference to FIG.

[0030] (3) Communications department 130 The communication unit 130 has a function of transmitting and receiving various information. The function of the communication unit 130 is realized by a communication device provided as hardware in the control device 13 of the abnormal sound detection device 10 described with reference to Fig. 1. In communication with the management server 20, the communication unit 130 receives an instruction to switch the detection model and transmits a notification (including notification data).

[0031] (4) Storage section 140 The storage unit 140 has a function of storing various information. The function of the storage unit 140 is realized by a storage medium provided as hardware by the control device 13 of the abnormal sound detection device 10 described with reference to Fig. 1, such as a hard disk drive (HDD), a solid state drive (SSD), a flash memory, an electrically erasable programmable read only memory (EEPROM), a random access read / write memory (RAM), a read only memory (ROM), or any combination of these storage media. As shown in FIG. 1, the storage unit 140 includes a detection model storage unit 141 and a detection data storage unit 142.

[0032] (4-1) Detection model storage unit 141 The detection model storage unit 141 has a function of storing a detection model. The detection model is a model capable of detecting drift noise from sounds collected in a target area. The detection model is a model (trained model) that has learned (machine learning) the characteristics of sounds when drift noise occurs (when abnormal sounds occur) and the characteristics of sounds when drift noise does not occur. The characteristics of sounds are shown, for example, by a spectrogram. In the spectrogram, frequency, time, and intensity are shown. A spectrogram image in which a spectrogram is visualized is used to generate (train) the detection model. In the spectrogram image, the spectrogram is shown as a three-dimensional graph with frequency on the vertical axis, time on the horizontal axis, and intensity on the color.

[0033] For the purpose of learning the detection model, multiple spectrogram images are prepared, which show that the spectrogram features in a specific frequency band are characteristics when drift noise is occurring (hereinafter also referred to as "spectrogram images when drift noise is occurring"), and multiple spectrogram images are prepared which show that the spectrogram features in a specific frequency band are characteristics when drift noise is not occurring (hereinafter also referred to as "spectrogram images when drift noise is not occurring"). The specific frequency band is, for example, around 4000 Hz. When a car is drifting, the frequency band around 4000 Hz contains the characteristics of the engine sound and muffler sound that occur during drifting. On the other hand, when a car is not drifting, the frequency band around 4000 Hz does not contain any particular sound characteristics. Therefore, in this embodiment, a spectrogram image showing a spectrogram of around 4000 Hz of sound collected during drift driving is used as the spectrogram image at generation time, and a spectrogram image showing a spectrogram of around 4000 Hz of sound collected during non-drift driving is used as the spectrogram image at non-generation time.

[0034] The detection model learns about the features of the spectrograms when drift noise is generated and when drift noise is not generated based on a data set consisting of a spectrogram image when drift noise is generated (first spectrogram image) and a spectrogram image when drift noise is not generated (second spectrogram image). For example, the detection model learns about the features of the spectrograms shown in each image around 4000 Hz. As a result, when a spectrogram image is input, the detection model can detect drift noise contained in the sound corresponding to the spectrogram image. For example, if the features around 4000 Hz of the input spectrogram image show similar features to the spectrogram image at occurrence, the detection model determines that the sound corresponding to the input spectrogram image contains drift noise. On the other hand, if the features around 4000 Hz of the input spectrogram image show similar features to the spectrogram image at non-occurrence, the detection model determines that the sound corresponding to the input spectrogram image does not contain drift noise.

[0035] In addition, characteristic sounds that occur during drift driving include the sound of tires sliding (squealing). In the spectrogram, the characteristics of the squealing sound appear in a frequency band of around 800 Hz. Therefore, the detection model can also detect drift noise by learning the characteristics of the frequency band of around 800 Hz of the spectrogram shown in the spectrogram image. However, the squealing sound may also occur during non-drift driving. Therefore, if the detection model learns only about the squealing sound, it may erroneously detect non-drift noise. Therefore, the detection model according to this embodiment learns at least about the characteristics of the spectrogram around 4000 Hz, and may also learn about the characteristics of the spectrogram around 800 Hz. If the detection model learns about both the characteristics of the spectrogram around 4000 Hz and around 800 Hz, the detection model can detect drift noise with higher accuracy.

[0036] Note that a number of different types of detection models may be prepared. For example, a detection model may be prepared for each type of weather. This is because the way sound travels changes depending on the weather. When detecting drift noise in a target area, the abnormal sound detection device 10 selects and uses a detection model that matches the weather at the time, thereby suppressing erroneous detection of drift noise and enabling more accurate detection of drift noise.

[0037] Here, the spectrogram image will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of a spectrogram image according to the present embodiment. Fig. 3 shows, as an example, a spectrogram image of a sound collected during drift driving.

[0038] In the spectrogram image 40 shown in FIG. 3, the vertical axis indicates frequency, the horizontal axis indicates time, and color indicates intensity. A region 41 in the spectrogram image 40 shows a spectrogram around 4000 Hz (3000 Hz to 5000 Hz). It can be seen that the width (horizontal axis direction) of the spectrogram is narrower around 4000 Hz compared to other frequency bands. A region 42 in the spectrogram image 40 shows a spectrogram around 800 Hz (600 Hz to 1000 Hz). It can be seen that the color of the spectrogram is different around 800 Hz compared to other frequency bands.

[0039] Here, a data set used for training the detection model will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of a data set used for training the detection model according to this embodiment.

[0040] As shown in Fig. 4, the data set 50 is composed of a spectrogram image 51 at the time of occurrence during drift driving and a spectrogram image 52 at the time of non-drift driving. Both spectrogram images are images in which spectrograms around 4000 Hz are extracted. The spectrogram image 51 at the time of occurrence shows the characteristics of drift noise around 4000 Hz. On the other hand, the spectrogram image 52 at the time of non-drift driving does not show the characteristics of drift noise around 4000 Hz.

[0041] (4-2) Detection data storage unit 142 The detection data storage unit 142 has a function of storing data related to the detection of drift noise. For example, the detection data storage unit 142 stores sound data to be detected, a spectrogram image of the sound data, the detection results of the drift noise, and the like.

[0042] (5) Control unit 150 The control unit 150 has a function of controlling the overall operation of the abnormal sound detection device 10. The functions of the control unit 150 are realized, for example, by causing a CPU (Central Processing Unit) or GPU (Graphics Processing Unit) provided as hardware in the control device 13 of the abnormal sound detection device 10 described with reference to FIG. As shown in FIG. 1, the control unit 150 includes a sound data acquisition unit 151, a spectrogram acquisition unit 152, an abnormal sound detection unit 153, and a notification control unit 154.

[0043] (5-1) Sound data acquisition unit 151 The sound data acquisition unit 151 has a function of acquiring sound data. For example, the sound data acquisition unit 151 acquires, as sound data, a sound collected by the sound collection device 11 (sound collection unit 110) provided in a target area where drift noise occurs. The sound data acquisition unit 151 outputs the acquired sound data to the spectrogram acquisition unit 152.

[0044] (5-2) Spectrogram acquisition unit 152 The spectrogram acquisition unit 152 has a function of performing spectrogram acquisition processing. For example, the spectrogram acquisition unit 152 acquires a spectrogram of a sound based on the sound data acquired by the sound data acquisition unit 151. Specifically, the spectrogram acquisition unit 152 acquires a spectrogram from the sound data, and generates a spectrogram image indicating the spectrogram. The spectrogram acquisition unit 152 outputs the generated spectrogram image to the abnormal sound detection unit 153.

[0045] (5-3) Abnormal sound detection unit 153 The abnormal sound detection unit 153 has a function of performing abnormal sound detection processing. The abnormal sound detection unit 153 detects an abnormal sound occurring in a target area at a predetermined time based on sound data of sounds collected at the predetermined time every time a predetermined time elapses. For example, the abnormal sound detection unit 153 detects an abnormal sound occurring at a predetermined time using a spectrogram image generated by the spectrogram acquisition unit 152 in predetermined time units based on the sound data acquired by the sound data acquisition unit 151. In this embodiment, the abnormal sound is drift noise generated by drifting of an automobile, so the abnormal sound detection unit 153 detects the drift noise based on the characteristics of the spectrogram around 4000 Hz where the characteristics of the drift noise appear.

[0046] Specifically, abnormal sound detection unit 153 detects an abnormal sound depending on whether the spectrogram characteristics at around 4000 Hz (a specific frequency band) shown in the spectrogram image (spectrogram) acquired by spectrogram acquisition unit 152 indicate characteristics when drift noise is occurring or characteristics when drift noise is not occurring. When the features around 4000 Hz in the spectrogram image indicate features when drift noise is occurring, the abnormal sound detection unit 153 determines that the sound corresponding to the spectrogram image contains drift noise and detects the drift noise.On the other hand, when the features around 4000 Hz in the spectrogram image indicate features when drift noise is not occurring, the abnormal sound detection unit 153 determines that the sound corresponding to the spectrogram image does not contain drift noise and does not detect the drift noise.

[0047] Note that the abnormal sound detection unit 153 according to this embodiment detects abnormal sounds using a detection model stored in the detection model storage unit 141. In this case, the abnormal sound detection unit 153 inputs the spectrogram image generated by the spectrogram acquisition unit 152 to the detection model, and acquires the detection result output from the detection model.

[0048] When multiple detection models of different types are prepared, the abnormal sound detection unit 153 detects the drift noise using an appropriate detection model. In this embodiment, a detection model is prepared for each type of weather. In this case, the abnormal sound detection unit 153 switches the detection model to be used depending on the weather when detecting the drift noise. This enables the abnormal sound detection unit 153 to obtain more accurate detection results depending on the weather at the time of detection. The determination of whether or not to switch the detection model is performed, for example, by the management server 20. Therefore, when the communication unit 130 receives an instruction to switch the detection model from the management server 20, the abnormal sound detection unit 153 switches the detection model in accordance with the switching instruction.

[0049] (5-4) Notification control unit 154 The notification control unit 154 has a function of controlling the notification of the detection result of the drift noise. For example, when the detection result of the drift noise at a predetermined time satisfies a predetermined condition, the notification control unit 154 notifies that an abnormal sound has been detected in the target area. The predetermined condition is set based on, for example, the number of times the drift noise is detected at a predetermined time and the duration of the detected drift noise. In the present embodiment, as an example, if drift noise is detected a predetermined number of times or for a predetermined duration in a predetermined time, the notification control unit 154 notifies that drift noise has been detected in the target area. For example, if drift noise is detected five times (an example of a predetermined number of times) or more within 30 seconds (an example of a predetermined time), the notification control unit 154 issues a notification. When notifying, the notification control unit 154 transmits a notification including notification data to the management server 20 via the communication unit 130.

[0050] <3. Functional configuration of the management server> The functional configuration of the abnormal sound detection device 10 according to this embodiment has been described above. Next, the functional configuration of the management server 20 according to this embodiment will be described with reference to Fig. 5. Fig. 5 is a block diagram showing an example of the functional configuration of the management server 20 according to this embodiment. As shown in FIG. 5, the management server 20 includes a communication unit 210, a storage unit 220, and a control unit 230.

[0051] (1) Communications unit 210 The communication unit 210 has a function of transmitting and receiving various information. In communication with the abnormal sound detection device 10, the communication unit 210 transmits an instruction to switch the detection model and receives notifications (including notification data). In communication with the receiving terminal 30, the communication unit 210 transmits notifications (including notification data) that drift noise has been detected, notifications that the timing of the drift noise occurrence has been predicted, aggregated data of the detection results, etc.

[0052] (2) Storage section 220 The storage unit 220 has a function of storing various types of information. The storage unit 220 is configured by a storage medium provided in the management server 20 as hardware, such as a HDD, SSD, flash memory, EEPROM, RAM, ROM, or any combination of these storage media. As shown in FIG. 5, the storage unit 220 includes a notification data storage unit 221, an aggregate data storage unit 222, and a weather information storage unit 223.

[0053] (2-1) Notification data storage unit 221 The notification data storage unit 221 has a function of storing notification data. The notification data storage unit 221 stores the notification data received by the communication unit 210 from the abnormal sound detection device 10.

[0054] (2-2) Aggregated data storage unit 222 The aggregated data storage unit 222 has a function of storing aggregated data. The aggregated data storage unit 222 stores aggregated data generated by the aggregation unit 232, which will be described later.

[0055] (2-3) Weather information storage unit 223 The weather information storage unit 223 has a function of storing weather information. The weather information storage unit 223 stores weather information acquired by the weather processing unit 235 described later. The weather information is information indicating the weather forecast for a target area.

[0056] (3) Control unit 230 The control unit 230 has a function of controlling the overall operation of the management server 20. The control unit 230 is realized, for example, by causing a CPU or a GPU that the management server 20 has as hardware to execute a program. As shown in FIG. 5, the control unit 230 includes a registration unit 231, a counting unit 232, a prediction unit 233, a linking unit 234, and a weather processing unit 235.

[0057] (3-1) Registration Section 231 The registration unit 231 has a function of registering various pieces of information. For example, the registration unit 231 stores and registers the notification data received by the communication unit 210 from the abnormal sound detection device 10 in the notification data storage unit 221.

[0058] (3-2) Tabulation section 232 The tallying unit 232 has a function of performing tallying processing. For example, the tallying unit 232 performs tallying on drift noise occurring in the target area. Specifically, the tallying unit 232 tallys up the occurrence of drift noise in the target area by occurrence time period, occurrence date, occurrence day of the week, and weather, based on the multiple notification data stored in the notification data storage unit 221, and generates tally data indicating the tallying results. The tallying unit 232 stores the generated tally data in the tally data storage unit 222. The tallying unit 232 also outputs the generated tally data to the prediction unit 233. The tallying unit 232 also transmits the tally data to the receiving terminal 30 via the communication unit 210.

[0059] (3-3) Prediction section 233 The prediction unit 233 has a function of performing prediction processing. For example, the prediction unit 233 predicts the timing of occurrence of drift noise based on the aggregated data aggregated by the aggregation unit 232. Specifically, the prediction unit 233 predicts that the drift noise will occur during a time period, a day, a day of the week, or in weather conditions that are associated with a high occurrence of drift noise, as indicated by the aggregated data. The prediction unit 233 transmits, via the communication unit 210, to the receiving terminal 30, a notification including the prediction result (notification data) of the occurrence timing of the drift noise.

[0060] (3-4) Collaboration Section 234 The linking unit 234 has a function of linking various types of data. For example, the linking unit 234 links a notification from the abnormal sound detection device 10 to the receiving terminal 30. In this embodiment, the abnormal sound is drift noise generated by drifting of a vehicle, so when the linking unit 234 is notified of the detection of drift noise from the abnormal sound detection device 10, it links with the receiving terminal 30 of the police or local government about the occurrence of drift noise in the target area. Linking with the receiving terminal 30 of the police or local government about the occurrence of drift noise means notifying the police or local government about the occurrence of drift noise. Note that the linking destination (notification destination) about the occurrence of drift noise is not limited to the receiving terminal 30 of the police or local government, and may be the receiving terminal 30 of a specified user.

[0061] In this way, in the abnormal sound detection system 1, when the abnormal sound detection device 10 detects the occurrence of drift noise, it notifies the management server 20, and the management server 20 coordinates the notification with the receiving terminal 30 of the police or local government. As a result, when the abnormal sound detection device 10 detects the occurrence of drift noise, the management server 20 notifies the police or local government in real time. Therefore, residents in the vicinity of the target area do not need to check for drift noise by themselves, and do not need to report it to the police or local government by telephone, etc. In addition, since the police or local government are notified in real time when drift noise is detected in the target area, there is no time lag from the occurrence of drift noise, and sufficient enforcement can be carried out.

[0062] (3-5) Weather Processing Unit 235 The weather processing unit 235 has a function of performing processing related to the weather. The weather processing unit 235 acquires weather information indicating a weather forecast for a target area from weather data acquired using, for example, an API (Application Programming Interface). The weather processing unit 235 stores the acquired weather information in the weather information storage unit 223. The weather processing unit 235 determines whether or not it is necessary to switch the detection model used by the abnormal sound detection device 10 based on the acquired weather information. For example, the weather processing unit 235 checks the timing at which the weather in the target area will change from the weather information. After checking, the weather processing unit 235 transmits a switching instruction to the abnormal sound detection device 10 via the communication unit 210 to switch to a detection model appropriate for the weather after the change at the timing at which the weather changes.

[0063] <4. Functional configuration of the receiving terminal> The functional configuration of the management server 20 according to this embodiment has been described above. Next, the functional configuration of the receiving terminal 30 according to this embodiment will be described with reference to Fig. 6 to Fig. 7. Fig. 6 is a block diagram showing an example of the functional configuration of the receiving terminal 30 according to this embodiment. As shown in FIG. 6, the receiving terminal 30 includes an input unit 310, a communication unit 320, a storage unit 330, a control unit 340, and an output unit 350.

[0064] (1) Input unit 310 The input unit 310 has a function of receiving input. The input unit 310 is configured, for example, by an input device that the receiving terminal 30 has as hardware, such as a button, a touch panel, a microphone, a mouse, a keyboard, and the like.

[0065] (2) Communications department 320 The communication unit 320 has a function of transmitting and receiving various information. In communication with the management server 20, the communication unit 320 receives a notification (including notification data) that a drift noise has been detected, a notification (including notification data) that the timing of occurrence of the drift noise has been predicted, and aggregated data of the detection results.

[0066] (3) Storage section 330 The storage unit 330 has a function of storing various information. The storage unit 330 is configured by a storage medium provided in the receiving terminal 30 as hardware, such as a HDD, SSD, flash memory, EEPROM, RAM, ROM, or any combination of these storage media. As shown in FIG. 6, the storage unit 330 includes an association data storage unit 331.

[0067] (3-1) Associated data storage unit 331 The linked data storage unit 331 has a function of storing data linked from the management server 20. For example, the linked data storage unit 331 stores notification data, aggregated data, and the like that the communication unit 320 receives from the management server 20.

[0068] (4) Control unit 340 The control unit 340 has a function of controlling the overall operation of the receiving terminal 30. The control unit 340 is realized, for example, by causing a CPU or a GPU that the receiving terminal 30 has as hardware to execute a program. As shown in FIG. 6, the control unit 340 includes a data acquisition unit 341 and an output processing unit 342.

[0069] (4-1) Data acquisition unit 341 The data acquisition unit 341 has a function of acquiring various data. For example, the data acquisition unit 341 acquires notification data and aggregated data that the communication unit 320 receives from the management server 20. The data acquisition unit 341 stores the acquired notification data and aggregated data in the linked data storage unit 331 and outputs the data to the output processing unit 342.

[0070] (4-2) Output processing unit 342 The output processing unit 342 has a function of controlling the output in the output unit 350. For example, the output processing unit 342 causes the output unit 350 to display the notification data and the tabulated data acquired by the data acquisition unit 341.

[0071] (5) Output unit 350 The output unit 350 has a function of outputting various information. The output unit 350 is, for example, an output device provided in the receiving terminal 30 as hardware, for example, a display device such as a display device or a touch screen (touch panel), or an audio output device such as a speaker. The output unit 350 displays notifications and aggregated data on a display device, for example, in response to control from the output processing unit 342.

[0072] Here, the tabulated data displayed on the receiving terminal 30 will be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of tabulated data displayed on the receiving terminal 30 according to this embodiment.

[0073] A management screen 60 shown in Fig. 7 displays various graphs generated based on the aggregated data. Area 61 displays the number of occurrences of drift noise aggregated by day. Area 62 displays the number of occurrences of drift noise aggregated by day of the week. Area 63 displays the number of occurrences of drift noise aggregated by time of day. Area 64 displays the number of occurrences of drift noise aggregated by weather. The management screen 60 may have a function that allows the user to select the location where the abnormal sound detection device 10 is installed (monitoring location) and the display period for the aggregated data.

[0074] <5. Processing flow> The functional configuration of the receiving terminal 30 according to this embodiment has been described above. Next, the flow of processing according to this embodiment will be described with reference to Figs.

[0075] (1) Flow of abnormal sound detection and notification The process flow of the abnormal sound detection process and the notification process will be described with reference to Fig. 8. Fig. 8 is a sequence diagram showing an example of the process flow of the abnormal sound detection process and the notification process in the abnormal sound detection system 1 according to this embodiment.

[0076] As shown in FIG. 8, first, the sound collection unit 110 of the abnormal sound detection device 10 collects sounds generated in a target area (step S101). The sound data acquisition unit 151 of the abnormal sound detection device 10 acquires sound data obtained by digitizing the sound collected by the sound collection unit 110 (step S102). The spectrogram acquisition unit 152 of the abnormal sound detection device 10 performs a spectrogram acquisition process (step S103). Specifically, the spectrogram acquisition unit 152 acquires a spectrogram from the sound data acquired by the sound data acquisition unit 151, and generates a spectrogram image showing the spectrogram.

[0077] If the predetermined time has elapsed (step S104 / YES), the process proceeds to step S105. On the other hand, if the predetermined time has not elapsed (step S104 / NO), the process repeats steps S101 to S103.

[0078] When the process proceeds to step S105, the abnormal sound detection unit 153 of the abnormal sound detection device 10 performs an abnormal sound detection process (step S105). After the abnormal sound detection process, the notification control unit 154 of the abnormal sound detection device 10 checks whether the detection result by the abnormal sound detection unit 153 satisfies a predetermined condition (step S106). If the predetermined condition is satisfied (step S106 / YES), the process proceeds to step S107. On the other hand, if the predetermined condition is not satisfied (step S106 / NO), the process repeats from step S101. If the process proceeds to step S107, the notification control unit 154 transmits a notification (including notification data) that drift noise has been detected in the target area to the management server 20 via the communication unit 130 (step S107).

[0079] The registration unit 231 of the management server 20 stores and registers the notification data received by the communication unit 210 from the abnormal sound detection device 10 in the notification data storage unit 221 (step S108). After the registration, the linking unit 234 of the management server 20 links (transmits) the notification data to the receiving terminal 30 via the communication unit 210 (step S109).

[0080] The data acquisition unit 341 of the receiving terminal 30 acquires the notification data received by the communication unit 320 from the management server 20 (step S110). The output processing unit 342 of the receiving terminal 30 causes the output unit 350 to display the notification data acquired by the data acquisition unit 341 (step S111).

[0081] (2) Processing flow for tabulation The flow of processing related to the tallying process will be described with reference to Fig. 9. Fig. 9 is a sequence diagram showing an example of the flow of processing related to the tallying process in the abnormal sound detection system 1 according to this embodiment.

[0082] As shown in FIG. 9, first, the counting unit 232 of the management server 20 acquires a plurality of pieces of notification data stored in the notification data storage unit 221 (step S201). The counting unit 232 performs counting processing based on the acquired plurality of pieces of notification data (step S202). After the tallying process, the tallying unit 232 transmits the tally data generated in the tallying process to the receiving terminal 30 via the communication unit 210 (step S203).

[0083] The data acquisition unit 341 of the receiving terminal 30 acquires the aggregated data received by the communication unit 320 from the management server 20 (step S204). The output processing unit 342 of the receiving terminal 30 causes the output unit 350 to display the tabulated data acquired by the data acquisition unit 341 (step S205).

[0084] (3) Flow of prediction process The flow of the prediction process will be described with reference to Fig. 10. Fig. 10 is a sequence diagram showing an example of the flow of the prediction process in the abnormal sound detection system 1 according to this embodiment.

[0085] As shown in FIG. 10, first, the prediction unit 233 of the management server 20 acquires a plurality of pieces of notification data stored in the notification data storage unit 221 (step S301). The prediction unit 233 performs a prediction process based on the acquired multiple pieces of notification data (step S302). After the prediction process, the prediction unit 233 transmits a notification including the prediction result obtained in the prediction process to the receiving terminal 30 via the communication unit 210 (step S303).

[0086] The data acquisition unit 341 of the receiving terminal 30 acquires the prediction result received by the communication unit 320 from the management server 20 (step S304). The output processing unit 342 of the receiving terminal 30 causes the output unit 350 to display the prediction result acquired by the data acquisition unit 341 (step S305).

[0087] (4) Weather processing flow The flow of processing related to weather processing will be described with reference to Fig. 11. Fig. 11 is a sequence diagram showing an example of the flow of processing related to weather processing in the abnormal sound detection system 1 according to this embodiment.

[0088] 11, first, the weather processing unit 235 of the management server 20 acquires weather information (step S401). At this time, the weather processing unit 235 may acquire the latest weather information from the Internet, or may acquire weather information that has been acquired in advance and stored in the weather information storage unit 223. The weather processor 235 checks whether or not there is a timing when the weather in the target area will change based on the acquired weather information (step S402). If there is a timing when the weather will change (step S402 / YES), the process proceeds to step S403. On the other hand, if there is no timing when the weather will change (step S402 / NO), the process ends. When the process proceeds to step S403, the weather processing unit 235 transmits an instruction to switch the detection model to the abnormal sound detection device 10 via the communication unit 210 (step S403).

[0089] The abnormal sound detection unit 153 of the abnormal sound detection device 10 switches the detection model used for detecting drift noise in accordance with the switching instruction received by the communication unit 130 from the management server 20 (step S404).

[0090] As described above, the abnormal sound detection system 1 of this embodiment comprises a sound data acquisition unit 151 that acquires, as sound data, sound collected by a sound collection device 11 provided in a target area where abnormal sound occurs, an abnormal sound detection unit 153 that detects abnormal sounds generated in the target area at predetermined times based on the sound data of sounds collected at predetermined times each time a predetermined time passes, and a notification control unit 154 that notifies that an abnormal sound has been detected in the target area if the detection result of the abnormal sound at the predetermined time satisfies predetermined conditions.

[0091] With this configuration, when drift noise occurs in the target area, it is detected by abnormal sound detection system 1, and the abnormal sound detection system 1 notifies the police and local government of the occurrence of drift noise. In other words, the occurrence of drift noise is reported to the police and local government in real time by abnormal sound detection system 1. This allows the occurrence of drift noise to be detected more accurately than if nearby residents were to check for the occurrence of drift noise, and eliminates the time lag that occurs between the occurrence of drift noise and the report when nearby residents manually report it. Therefore, the abnormal sound detection system 1 according to this embodiment makes it possible to grasp the occurrence of drift noise (abnormal sound) more accurately and in real time.

[0092] <6. Variations> The embodiments of the present invention have been described above. Next, modified examples of the embodiments of the present invention will be described. Each modified example described below may be applied alone to the embodiments of the present invention, or may be applied in combination to the embodiments of the present invention. Furthermore, each modified example may be applied in place of the configuration described in the embodiments of the present invention, or may be applied in addition to the configuration described in each embodiment of the present invention.

[0093] The abnormal sound detection system 1 according to the embodiment described above may have a photographing function. In this case, in the configuration of the abnormal sound detection system 1 shown in FIG. 1, the abnormal sound detection device 10 further includes a photographing device. Like the sound collection device 11, the photographing device is provided outside the housing of the abnormal sound detection device 10 and connected to the control device 13 inside the housing of the abnormal sound detection device 10. Furthermore, in the functional configuration of the abnormal sound detection device 10 shown in FIG. 2, the abnormal sound detection device 10 further includes a photographing unit. The function of the photographing unit is realized by the photographing device that the abnormal sound detection device 10 includes as hardware. The photographing device (camera) is installed facing the target area so as to be able to photograph the target area. The photographing device is connected to the control device 13 by wired or wireless connection so as to be able to communicate with the control device 13. The photographing device photographs an image (still image or moving image) of the target area, for example, in accordance with a control signal received from the control device 13. After photographing an image of the target area, the photographing device transmits the photographed image to the control device 13. For example, when drift noise is detected by the abnormal sound detection unit 153 of the control device 13, the control signal is generated by the abnormal sound detection unit 153 and transmitted to the photographing device. The photographing device photographs the target area when triggered by receiving a control signal from the control device 13, i.e., when drift noise is detected in the target area. In this way, the camera does not take pictures all the time, but only takes pictures when it receives a triggering control signal. Of the images taken, only images taken before and after the occurrence of drift noise or images taken after the occurrence are saved. This reduces data volume and power consumption, and enables longer-term monitoring.

[0094] The notification control unit 154 of the abnormal sound detection device 10 may also transmit an image captured by the imaging device when sending a notification to the management server 20 via the communication unit 130. Furthermore, the linking unit 234 of the management server 20 may also link the image received from the abnormal sound detection device 10 when linking with the receiving terminal 30 about the occurrence of drift noise in the target area. The output processing unit 342 of the receiving terminal 30 causes the output unit 350 to display the image together with the notification data received from the management server 20. This allows the user of the receiving terminal 30 to obtain information such as the number and characteristics of automobiles that are generating drift noise in the target area. The characteristics of the automobiles include, for example, the model, body color, license plate number, etc.

[0095] In the above-described embodiment, an example in which the abnormal sound is drift noise has been described, but the present invention is not limited to such an example. The abnormal sound may be, for example, a mechanical sound generated when a malfunction occurs in a machine installed in a factory or the like. In this case, the target area is an area in a factory or the like in which the target machine is installed. That is, in the target area, a state in which a malfunction occurs in a machine is an abnormal state, and a mechanical sound generated in that state is an abnormal sound. On the other hand, a state in which no malfunction occurs in a machine in the target area is a normal state, and a mechanical sound generated in that state is a normal sound. When the abnormal sound is a mechanical sound, the detection model learns the characteristics of the mechanical sound (abnormal sound) generated when a malfunction occurs in a target machine installed in the target area, and the characteristics of the mechanical sound (normal sound) generated when the target machine is not malfunctioning. Through this learning, the detection model can detect abnormal sounds generated from machines installed in the target area. The abnormal sound detection device 10 detects abnormal sounds using the detection model learned in this way, and issues a notification when an abnormal sound is detected. The notification destination may be, for example, a terminal used by a worker or the like installed at a site such as a factory, or a terminal used by a manager or the like installed at a location other than the site. This allows the worker or manager to know in real time that a malfunction has occurred in a machine installed at a factory, etc., and to quickly respond to the malfunction of the machine.

[0096] Furthermore, in the above-described embodiment, an example has been described in which the abnormal sound detection device 10 has the function of detecting abnormal sounds, but the present invention is not limited to this example. The function of detecting abnormal sounds using a detection model (the function of the abnormal sound detection unit 153) may be possessed by the management server 20. In this case, the function of acquiring a spectrogram may be possessed by either the abnormal sound detection device 10 or the management server 20. The function of controlling notification of abnormal sound detection (the function of the notification control unit 154) is possessed by the management server 20, and controls notification to the receiving terminal 30.

[0097] Furthermore, the abnormal sound detection unit 153 may have a function of switching whether or not to detect an abnormal sound depending on the weather or time of day when the abnormal sound is detected. In this case, the abnormal sound detection unit 153 does not detect an abnormal sound, for example, in weather conditions that reduce the accuracy of sound collection. Examples of weather conditions that reduce the accuracy of sound collection include rainy weather and windy weather. Furthermore, the abnormal sound detection unit 153 may perform the abnormal sound detection process only during, for example, nighttime hours.

[0098] The abnormal sound detection system 1 may also be configured to suppress erroneous detection of abnormal sounds. For example, the abnormal sound detection process may be performed only during the nighttime, or collected sounds whose volume is equal to or lower than a predetermined value may be excluded from detection of abnormal sounds. The setting values ​​configured for these settings may be variable depending on the location where the abnormal sound detection device 10 is installed.

[0099] The above describes the modified embodiment of the present invention. In addition, some or all of the functions of the abnormal sound detection system 1, the abnormal sound detection device 10, the management server 20, and the receiving terminal 30 in the above-mentioned embodiment may be realized by a computer. In that case, a program for realizing the functions may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read into a computer system and executed to realize the functions. In addition, the term "computer system" as used herein includes hardware such as an OS and peripheral devices. In addition, the term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into a computer system. Furthermore, the term "computer-readable recording medium" may include a medium that dynamically holds a program for a short period of time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. In addition, the above-mentioned program may be a program for realizing some of the above-mentioned functions, or may be a program that can realize the above-mentioned functions in combination with a program already recorded in the computer system, or may be a program that is realized using a programmable logic device such as an FPGA (Field Programmable Gate Array).

[0100] Although the embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design changes, etc. are possible within the scope that does not deviate from the gist of the present invention. [Explanation of symbols]

[0101] 1...abnormal sound detection system, 10...abnormal sound detection device, 11...sound collection device, 12...power supply device, 13...control device, 20...management server, 30...receiving terminal, 110...sound collection unit, 120...power supply unit, 130...communication unit, 140...storage unit, 141...detection model storage unit, 142...detection data storage unit, 150...control unit, 151...sound data acquisition unit, 152...spectrogram acquisition unit, 153...abnormal sound detection unit, 154...notification control unit, 210... Communication unit, 220... memory unit, 221... notification data memory unit, 222... aggregated data memory unit, 223... weather information memory unit, 230... control unit, 231... registration unit, 232... aggregation unit, 233... prediction unit, 234... linkage unit, 235... weather processing unit, 310... input unit, 320... communication unit, 330... memory unit, 331... linked data memory unit, 340... control unit, 341... data acquisition unit, 342... output processing unit, 350... output unit, NW... network

Claims

1. a sound data acquisition unit that acquires, as sound data, sounds collected by a sound collection device provided in a target area where abnormal sounds occur; an abnormal sound detection unit that detects the abnormal sound generated in the target area at a predetermined time based on the sound data of the sound collected at the predetermined time every time a predetermined time elapses; a notification control unit that notifies the user that an abnormal sound has been detected in the target area when a result of the abnormal sound detection at the specified time satisfies a specified condition; and An abnormal sound detection system comprising:

2. a spectrogram acquisition unit that acquires a spectrogram of the sound based on the acquired sound data; Further equipped with the abnormal sound detection unit detects the abnormal sound depending on whether a spectrogram feature in a specific frequency band indicated by the acquired spectrogram indicates a feature when an abnormal sound is occurring or a feature when no abnormal sound is occurring. The abnormal sound detection system according to claim 1 .

3. When the abnormal sound is a drift sound generated by a drifting of a vehicle, The abnormal sound detection unit detects the noise based on spectrogram characteristics around 4000 Hz, where characteristics of the drift sound appear. The abnormal sound detection system according to claim 2 .

4. the abnormal sound detection unit detects the abnormal sound using a detection model that has learned about spectrogram features when an abnormal sound is occurring and when an abnormal sound is not occurring, based on a data set consisting of a first spectrogram image that indicates that spectrogram features in the specific frequency band are features when an abnormal sound is occurring, and a second spectrogram image that indicates that spectrogram features in the specific frequency band are features when an abnormal sound is not occurring. The abnormal sound detection system according to claim 2 .

5. The detection model is prepared for each type of weather, the abnormal sound detection unit switches the detection model to be used depending on the weather when detecting the abnormal sound; The abnormal sound detection system according to claim 4 .

6. the abnormal sound detection unit switches whether or not to detect the abnormal sound depending on the weather or the time of day when the abnormal sound is detected. The abnormal sound detection system according to claim 1 .

7. The notification control unit notifies the user that the abnormal sound has been detected in the target area when the abnormal sound has been detected a predetermined number of times or for a predetermined duration or longer within the predetermined time period. The abnormal sound detection system according to claim 1 .

8. When the abnormal sound is a drift sound generated by a drifting of a vehicle, a linking unit that, when notified of the detection of the abnormal sound, links to a predetermined user that the abnormal sound is occurring in the target area; The abnormal sound detection system according to claim 1 , further comprising:

9. a counting unit that, based on notification data indicating data related to the abnormal sound that has been notified, counts the occurrence of the abnormal sound in the target area by time period of occurrence, date of occurrence, day of the week of occurrence, and weather, and generates counted data indicating the counting results; The abnormal sound detection system according to claim 1 , further comprising:

10. a prediction unit that predicts an occurrence timing of the abnormal sound based on the collected data; The abnormal sound detection system according to claim 9 , further comprising:

11. a sound data acquisition process for acquiring, as sound data, sounds collected by a sound collection device provided in a target area where abnormal sounds occur; an abnormal sound detection process for detecting the abnormal sound generated in the target area at a predetermined time based on the sound data of the sound collected at the predetermined time, every time a predetermined time elapses; a notification control step of notifying the user that the abnormal sound has been detected in the target area when the abnormal sound detection result at the specified time satisfies a specified condition; An abnormal sound detection method implemented by a computer, comprising:

12. Computer, a sound data acquisition means for acquiring, as sound data, sounds collected by a sound collection device provided in a target area where abnormal sounds are generated; an abnormal sound detection means for detecting the abnormal sound generated in the target area at a predetermined time based on the sound data of the sound collected at the predetermined time each time a predetermined time elapses; a notification control means for notifying the user that an abnormal sound has been detected in the target area when the abnormal sound detection result at the predetermined time satisfies a predetermined condition; A program to function as a

Images