Emergency notification system and its program

Abstract

The present disclosure provides an emergency notification system and a program therefor that function effectively even in situations where a user is unable to recognize the situation and notify an appropriate contact. [Solution] Emergency notification systems 10a to 10c according to the present disclosure are emergency notification systems that automatically notify users of their status. Each emergency notification system 10a to 10c includes a sound collection device that collects the user's voice and sounds around the user, a first location information device that acquires location information, a storage device that stores the sound data collected by the sound collection device and the first location information acquired by the first location information device, a calculation device that estimates the user's status based on the sound data and the first location information and selects multiple contacts according to the estimated user's status, and a communication device that automatically notifies the multiple contacts selected by the calculation device using preset communication means.

Description

[Technical Field]

[0001] The present disclosure relates to an emergency notification system that automatically makes a notification depending on the user's situation, and a program therefor. [Background technology]

[0002] Conventionally, when a user encounters an emergency situation such as an incident, accident, crime, sudden illness, fall, or vehicle theft, they have had to press a button on a portable reporting device or operate an app on a portable communication device to report to the police, etc. Furthermore, Japanese Patent Laid-Open Publication No. 2003-217058 (Patent Document 1) discloses an emergency information transmission device that responds quickly to an emergency situation and helps to mitigate damage, which recognizes registered voice identification data and transmits predetermined information based on the registered voice identification data to a destination via a communication line. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2003-217058 Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional notification devices and portable communication devices require active operation by the user, such as pressing a button or starting an application, and have the problem of not functioning effectively when such operation is not possible. Also, the emergency information transmission device disclosed in Japanese Patent Laid-Open No. 2003-217058 (Patent Document 1) is capable of notifying contacts by voice, but requires the user to recognize the situation and issue a voice password, which has the problem of not functioning effectively when the user is unable to recognize the situation, such as when the user loses consciousness.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and its purpose is to provide an emergency notification system and a program therefor that function effectively even in situations where the user is unable to recognize the situation and notify the appropriate contact. [Means for solving the problem]

[0006] The emergency call system according to the present disclosure is an emergency call system that automatically makes a call depending on the user's situation. The emergency call system includes: a sound collection device that collects the user's voice and sounds around the user; No. 1 A first location information device that acquires location information and sound data collected by a sound collection device Ta a storage device for storing the sound data and the first position information; At least sound data Based on this, the system estimates the user's situation and provides multiple contacts according to the estimated user situation. Search storage devices or networks A calculation device that performs search and a communication device that automatically notifies the plurality of contacts by a preset communication means.

[0007] The program according to the present disclosure includes a sound collection device that collects a user's voice and sounds around the user, and No. 1 A first location information device that acquires location information and sound data collected by a sound collection device Ta The system includes a storage device for storing the data, a computing device, and a communication device. In the system , automatically reports based on the user's situation for It is a program. For the computing device, Sound data and first location information At least sound data Based on this, the system estimates the user's situation and then searches for multiple contacts according to the estimated user situation. Search storage devices or networks and search a process of automatically notifying the plurality of contacts by a communication means of a communication device that is set in advance; The program to run . [Effects of the Invention]

[0008] The emergency call system according to the present disclosure estimates the user's situation based on sound data and first location information and selects multiple contacts according to the estimated user's situation, so that even if the user is unable to recognize the situation, it can automatically notify appropriate contacts and effectively function as an emergency call system. As a result, the emergency call system according to the present disclosure can contribute to ensuring the safety of children, the elderly, people with disabilities, people requiring care, patients, etc., and can also respond quickly to situations such as vehicle theft, abuse, and sudden illness. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a schematic diagram illustrating an example of a situation in which a user automatically makes a report using an emergency report system according to the present disclosure. [Figure 2] 1 is a schematic diagram illustrating a hardware configuration of an emergency call system according to the present disclosure. [Figure 3] FIG. 10 is a diagram illustrating an example of sound data collected by a microphone of the emergency call system according to the present disclosure. [Figure 4] FIG. 2 is a block diagram for explaining the processing of sound data in the emergency notification system according to the present disclosure. [Figure 5] 10 is a flowchart for explaining the processing of the emergency call system according to the present disclosure. [Figure 6] FIG. 10 is a schematic diagram illustrating a configuration of a modified example of the emergency notification system according to the present disclosure. [Figure 7] FIG. 1 is a diagram for explaining an estimation model of an emergency notification system according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference numerals and description thereof will not be repeated. (Embodiment) [Emergency notification system] The emergency notification system according to the present disclosure can estimate emergency situations such as incidents, accidents, crimes, sudden illness, falls, and vehicle theft based on sound data and location information, and automatically notify appropriate contacts. The emergency notification system can be applied to a variety of emergency situations, but the following will specifically explain the situation of sudden illness and accidents as examples. FIG. 1 is a schematic diagram showing an example of a user situation in which a notification is automatically made by the emergency notification system according to the present disclosure. Situation A shown in FIG. 1 is, for example, a situation in which a person requiring care (user) falls suddenly ill in a nursing home. Meanwhile, situation B is, for example, a situation in which a child (user) is hit by a vehicle 22 at an intersection.

[0011] In situation A, emergency notification system 10a is installed, for example, in a nursing bed 20, and detects the groans, sounds of falling, etc. of the person requiring care (user) and location information to estimate an emergency situation and automatically notify appropriate contacts (for example, a hospital, family, staff, etc.). In situation B, emergency notification system 10b is installed, for example, in a school bag 21 of a child (user), and detects the child's (user's) voice, surrounding sounds, etc. and location information to estimate an emergency situation and automatically notify appropriate contacts (for example, a fire department, family, etc.). Also in situation B, emergency notification system 10c is installed, for example, in the rearview mirror, dashboard, etc. of a vehicle 22, and detects the driver's (user's) voice, surrounding sounds, etc. and location information to estimate an emergency situation and automatically notify appropriate contacts (for example, the police, a company, etc.).

[0012] The hardware configuration of the emergency notification systems 10a to 10c will be described. Fig. 2 is a schematic diagram showing the hardware configuration of the emergency notification system 10a according to the present disclosure. The emergency notification system 10a includes a computing device 11, a main memory 12, an input device 13, a display 14, a storage device 15, a communication device 16, a microphone 17 (a sound collection device), a camera 18 (an image capture device), and a GPS 19 (a first position information device). The emergency notification systems 10b and 10c include the same hardware configuration as the emergency notification system 10a, and therefore detailed description thereof will not be repeated.

[0013] Since the emergency notification system 10a is installed in the nursing care bed 20, all components, including the computing device 11, are implemented in a single housing. However, the emergency notification system may be implemented in a single housing like the emergency notification system 10a, or in multiple housings, as long as they can cooperate to execute processing. For example, in the emergency notification system 10b installed in the school bag 21, the communication device 16, microphone 17, and GPS 19 may be implemented in the school bag 21, while the other components may be implemented in a terminal device such as a smartphone owned by the user or the user's family. In the emergency notification system 10c installed in the vehicle 22, the microphone 17, camera 18, and GPS 19 installed in the vehicle 22 and other components using these components may be implemented in separate housings. In addition to the nursing care bed 20, school bag 21, and vehicle 22, the emergency notification system may also be implemented in, for example, a smart key, a wallet, a writing instrument, or a wearable device. The emergency call system can change the chassis to which the hardware configuration is allocated depending on the items to be implemented.

[0014] In the emergency notification system 10a, the arithmetic device 11 is configured, for example, by a CPU (Central Processing Unit), an FPGA (Field Programmable Gate Array), a GPU (Graphics Processing Unit), or an MPU (Multi-Processing Unit). The arithmetic device 11 may also be configured by a processing circuit. The arithmetic device 11 can read out programs (for example, an OS 151 and a control program 152) stored in the storage device 15, load them into the main memory 12, and execute them.

[0015] By executing the control program 152 stored in the storage device 15, the arithmetic device 11 estimates the user's situation based on the sound data collected by the microphone 17 and the location information acquired by the GPS 19, and performs a process of selecting multiple contacts according to the estimated user's situation. Furthermore, the arithmetic device 11 performs a process of notifying the selected contacts using the communication device 16.

[0016] The main memory 12 is configured with a volatile storage device such as a dynamic random access memory (DRAM) or a static random access memory (SRAM). The storage device 15 is configured with a nonvolatile storage device such as a hard disk drive (HDD) or a solid state drive (SSD). In addition to an OS 151 for realizing basic functions, the storage device 15 stores a control program 152, sound data collected by a microphone 17, still images and videos captured by a camera 18, and the like.

[0017] The input device 13 includes not only a keyboard and a mouse connected to the emergency call system 10a, but also devices capable of receiving user operations, such as a touch panel and operation buttons provided on the surface of the display 14. The input device 13 can be set using a GUI (Graphical User Interface) function so that the user can easily play back audio data, videos, and the like recorded in the storage device 15.

[0018] Display 14 is a flat panel display such as an LCD (Liquid Crystal Display) or an organic EL (Electroluminescence) display. Display 14 can display information based on sound data collected by microphone 17 and location information acquired by GPS 19, as well as still images and videos captured by camera 18.

[0019] The communication device 16 supports multiple communication methods, such as mobile communication (4G, 5G, etc.), wireless LAN (Local Area Network), and Bluetooth (registered trademark). Therefore, the communication device 16 can communicate with the outside world using multiple communication methods, such as voice calls, emails, SMS, and app notifications. The communication device 16 can also transmit audio data collected by the microphone 17 and still images and videos (imaging data) captured by the camera 18 to an external server or a cloud server. From the perspective of privacy and confidentiality protection, the communication device 16 may encrypt the audio data collected by the microphone 17 and the still images and videos (imaging data) captured by the camera 18 before transmitting them to the external server or the cloud server. Furthermore, the communication device 16 may download the control program 152 and other programs from another server to the storage device 15.

[0020] Microphone 17 can collect the user's voice and sounds around the user. Microphone 17 may be provided in a housing different from that of arithmetic device 11 and the like, and may be configured to transmit sound data to the housing including arithmetic device 11 and the like via a wired or wireless connection. Camera 18 can capture still images and videos of the user and / or the user's surroundings. Camera 18 may be provided in a housing different from that of arithmetic device 11 and the like, and may be configured to transmit still image and video data to the housing including arithmetic device 11 and the like via a wired or wireless connection.

[0021] The GPS 19 is a GPS (Global Positioning System) receiver and can acquire location information. The GPS 19 may be provided in a housing different from the arithmetic device 11 and the like, and may be configured to transmit the location information to the housing including the arithmetic device 11 and the like via a wired or wireless connection. [Emergency notification system operation] Next, the operation of the emergency notification systems 10a to 10c will be described. FIG. 3 is a diagram showing an example of sound data collected by a microphone of the emergency notification system according to the present disclosure. First, the operation of the emergency notification system 10a in a situation A in which a person requiring care (user) suddenly falls ill and collapses in the nursing facility shown in FIG. 1 will be described. For example, if the person requiring care (user) feels unwell and utters "help me" before suddenly falling ill and collapses next to the nursing bed 20, the microphone 17 of the emergency notification system 10a will collect the sound data shown in FIG. 3(A). Note that the sound data shown in FIG. 3(A) is illustrated with the horizontal axis representing time and the vertical axis representing sound pressure, but the sound data also includes information such as sound frequency in addition to sound pressure.

[0022] When the arithmetic device 11 receives sound data such as that shown in FIG. 3A from the microphone 17, it performs processing to analyze the sound pressure, sound frequency, and speech content from the sound data. FIG. 4 is a block diagram for explaining processing of sound data in the emergency call system according to the present disclosure. First, the arithmetic device 11 performs input processing 171 to receive sound data in real time, including the user's voice (e.g., screaming) and sounds around the user (e.g., the sound of falling) collected by the microphone 17. The arithmetic device 11 then performs A / D conversion processing 172 to convert the analog sound data received by the input processing 171 into digital sound data. Note that when outputting sound data from the microphone 17 to the arithmetic device 11, if the microphone 17 converts the digital sound data and outputs it, the arithmetic device 11 skips the A / D conversion processing 172.

[0023] Next, the arithmetic device 11 performs signal preprocessing 173 such as noise removal and normalization on the digital sound data. Note that if the digital sound data has sound quality that can be sufficiently processed in subsequent processing, the arithmetic device 11 may skip the signal preprocessing 173. Furthermore, the arithmetic device 11 performs voice analysis processing 174 on the digital sound data. In the voice analysis processing 174, voice feature extraction processing and natural language processing (NLP) are performed on the voice included in the sound data to extract pre-registered specific keywords (e.g., "help," "ouch," "ah," "ahh," etc.). Note that the arithmetic device 11 may register the user's voice features (e.g., voiceprint, etc.) in advance so that it can identify whether the person who uttered the specific keywords is the user or another person.

[0024] Next, when the arithmetic device 11 detects a sound pressure level equal to or higher than a standard level included in the sound data or a specific keyword registered in advance included in the sound data, the arithmetic device 11 starts a situation estimation process 175 that estimates the user's situation. The situation estimation process 175 estimates the user's situation based on the sound data and location information (first location information) acquired by the GPS 19. The situation estimation process 175 determines whether the user's situation is an emergency situation based on rules, for example, based on the sound pressure level equal to or higher than a standard level, the sound frequency, the voice containing the specific keyword, and the location information acquired by the GPS 19.

[0025] Specifically, in the case of the sound data shown in FIG. 3(A), the situation estimation process 175 starts processing from time t1 when the specific keyword "help me" is uttered, and determines the user's situation by taking into account sound data from the period before and after time t1. Note that in the case of the sound data shown in FIG. 3(A), the utterance of the specific keyword "help me" is the trigger sound. The situation estimation process 175 determines that an emergency situation has occurred in which the person requiring care (user) has suddenly fallen ill, because there is no abnormal sound during the period R1 from time t1 to time t0, and a loud "bang" sound is emitted at time t2 after time t1. Note that the emergency notification system 10a always stores sound data for a certain period in the storage device 15 so that sound data from the period before and after time t1 can be taken into account. Furthermore, in the situation estimation process 175, an example has been described in which the user's situation is estimated by taking into account sound data from a period before and after time t1. However, the user's situation may be estimated only from the sound data and position information from time t1, without taking into account sound data from a period before and after time t1.

[0026] The arithmetic device 11 selects multiple contacts according to the user's situation estimated in the situation estimation process 175, and performs output process 176 to output the results to the communication device 16. Specifically, when the arithmetic device 11 determines in the situation estimation process 175 that the situation is an emergency situation in which the care recipient (user) has suddenly fallen ill, the arithmetic device 11 selects family members, staff members, and a fire department for calling an ambulance as pre-set contacts, and outputs the selected contacts to the communication device 16. The communication device 16 automatically notifies the multiple contacts selected by the arithmetic device 11 by pre-set communication means. Specifically, as shown in FIG. 1, the communication device 16 of the emergency notification system 10a notifies family members and staff members by email that the care recipient (user) has suddenly fallen ill, and simultaneously notifies the fire department by phone to call an ambulance. Note that in an emergency situation in which the care recipient (user) has suddenly fallen ill, the care recipient (user) may not be able to call the fire department himself / herself. In such a case, the call to the fire department may be forwarded to a family member or staff member who received the email notification.

[0027] In the example shown in FIG. 3(A), a notification is initiated at time t3. The period R2 from time t1 to time t3 is the period during which the situation estimation process 175 estimates the user's situation, which is, for example, one to several seconds. Therefore, the emergency notification system 10a can perform the process from detecting an emergency situation to issuing a notification in almost real time. Furthermore, as shown in FIG. 3(A), the emergency notification system 10a starts recording audio using the microphone 17 or / and recording audio using the camera 18 from time t1. The audio data recorded by the microphone 17 or / and recording audio using the camera 18 is transmitted to an external server or cloud server via the communication device 16. This allows for recording of evidence of an emergency situation in which the care recipient (user) has suddenly fallen ill. The audio data recorded by the microphone 17 or / and recording audio using the camera 18 is continued for a certain period (e.g., 30 seconds) from the start. The emergency notification system 10a installed in the nursing bed 20 can respond not only to sudden illnesses of the care recipient but also to emergencies such as the care recipient's fall or wandering, thereby ensuring the care recipient's peace of mind and safety.

[0028] Next, the operation of emergency notification systems 10b and 10c in situation B where a child (user) is hit by vehicle 22 at the intersection shown in Fig. 1 will be described. For example, if a loud "bang" is emitted when a child (user) is hit by vehicle 22, microphone 17 of emergency notification system 10b collects sound data shown in Fig. 3(B). Note that the sound data shown in Fig. 3(B) is illustrated with time on the horizontal axis and sound pressure on the vertical axis, but the sound data also includes sound frequency and the like in addition to sound pressure.

[0029] In the case of the sound data shown in FIG. 3(B), the situation estimation process 175 starts processing from time t1 when a "bang" sound with a sound pressure level equal to or higher than the reference level is emitted, and determines the user's situation by taking into account sound data from a period before and after time t1. In the case of the sound data shown in FIG. 3(B), the "bang" sound with a sound pressure level equal to or higher than the reference level is the trigger sound. The situation estimation process 175 determines that an emergency situation has occurred in which a child (user) has been hit by a vehicle 22, based on the fact that a "squeak" sound of the vehicle 22 braking is emitted during a period R1 from time t1 to time t0, and the specific keyword "ouch" is uttered at time t2 after time t1. The situation estimation process 175 may also take into account location information from the GPS 19 of the emergency notification system 10b to determine that an emergency situation has occurred in which a child (user) has been hit by a vehicle 22. The emergency notification system 10b always stores sound data for a certain period in the storage device 15 so that sound data from a period before and after time t1 can be taken into account.

[0030] When the calculation device 11 determines in the situation estimation process 175 that the situation is an emergency situation in which the child (user) has been hit by the vehicle 22, it selects the family and the fire department to call an ambulance as pre-set contacts and outputs them to the communication device 16. As shown in FIG. 1, the communication device 16 of the emergency notification system 10b notifies the family by email that the child (user) has been hit by the vehicle 22, and simultaneously notifies the fire department by phone to call an ambulance. Note that in an emergency situation in which the child (user) has been hit by the vehicle 22, the child (user) may not be able to call the fire department himself / herself, and in that case, the call to the fire department may be forwarded to the family who has received the email notification.

[0031] In the example shown in FIG. 3(B), a notification is initiated at time t3, and the period R2 from time t1 to time t3 is the period during which the situation estimation process 175 estimates the user's situation, which is, for example, one to several seconds. Therefore, the emergency notification system 10b can perform the process from detecting an emergency situation to reporting in almost real time. Furthermore, as shown in FIG. 3(B), the emergency notification system 10b starts recording audio using the microphone 17 and / or the camera 18 from time t1. The data recorded by the microphone 17 and / or the camera 18 is transmitted to an external server or a cloud server via the communication device 16. This allows evidence of an emergency situation in which a child (user) has been hit by a vehicle 22 to be recorded.

[0032] 1, not only is emergency notification system 10b provided in school bag 21, but emergency notification system 10c is also provided in vehicle 22. Therefore, emergency notification system 10c can also estimate an emergency situation in which vehicle 22 has hit a child from sound data and location information. Although sound data used by emergency notification system 10c to determine an emergency situation in which vehicle 22 has hit a child is not shown, processing begins from the time when a "bang" sound with a sound pressure level above a certain level is emitted, and the situation of the driver (user) can be determined taking into account sound data from the period before and after that time.

[0033] If the situation estimation process 175 determines that the situation is an emergency situation in which the vehicle 22 has struck a child, the calculation device 11 of the emergency notification system 10c selects the insurance company and the police as pre-set contacts and outputs them to the communication device 16. As shown in FIG. 1, the communication device 16 of the emergency notification system 10c notifies the insurance company of the accident involving the vehicle 22 and simultaneously notifies the police by telephone. Note that the communication device 16 of the emergency notification system 10c may also notify the fire department by telephone to call an ambulance in addition to the police. Furthermore, the emergency notification system 10c starts recording with the microphone 17 and / or the camera 18 from the time when a "bang" sound with a sound pressure level above a certain level is emitted. The data recorded by the microphone 17 and / or the camera 18 is transmitted to an external server or a cloud server via the communication device 16. This allows evidence of the emergency situation in which the vehicle 22 struck a child to be recorded.

[0034] The emergency notification system 10b provided in the school bag 21 can respond not only to accidents such as being hit by the vehicle 22, but also to emergency situations such as kidnapping, ensuring the safety and security of the child (user). The emergency notification system 10c provided in the vehicle 22 can respond not only to accidents such as being hit by a vehicle, but also to emergency situations such as theft of the vehicle 22, ensuring the safety of the driver (user) and the safety of the vehicle 22. [Emergency notification system flowchart] Next, the processing of the emergency notification systems 10a to 10c will be described using a flowchart. Fig. 5 is a flowchart for explaining the processing of the emergency notification systems 10a to 10c according to the present disclosure. First, the arithmetic device 11 of the emergency notification systems 10a to 10c collects sound data for a certain period of time using the microphone 17 and stores the data in the storage device 15 (step S101). The arithmetic device 11 determines whether or not a trigger sound has been detected from the sound data (step S102). Here, the trigger sound is a sound with a sound pressure above a reference level, a sound containing a specific keyword, or the like.

[0035] If a trigger sound has not been detected (NO in step S102), the arithmetic device 11 returns the process to step S101. If a trigger sound has been detected (YES in step S102), the arithmetic device 11 starts estimating the user's situation (step S103). The arithmetic device 11 starts capturing images with the camera 18 (step S104). The arithmetic device 11 acquires sound data before and after the trigger sound from the storage device 15 (step S105). The arithmetic device 11 acquires location information from the GPS 19 (step S106).

[0036] The arithmetic device 11 determines whether to estimate the user's situation based on the trigger sound, the sound data acquired in step S105, and the location information acquired by the GPS 19 (step S107). If sufficient data has not yet been acquired to estimate the user's situation (NO in step S107), the arithmetic device 11 returns the process to step S105. If sufficient data has been acquired to estimate the user's situation (YES in step S107), the arithmetic device 11 estimates the user's situation and selects a contact based on the user's situation (step S108).

[0037] The arithmetic unit 11 determines whether the selected contact has been preset (step S109). If the selected contact has not been preset (NO in step S109), the arithmetic unit 11 searches for the contact from the storage device 15 and / or the network (step S110). If the selected contact has been preset (YES in step S109), the arithmetic unit 11 skips the process of step S110.

[0038] The arithmetic device 11 contacts the selected contacts in sequence and notifies them of the user's status (step S111). The arithmetic device 11 determines whether or not all contacts have been contacted (step S112). If not all contacts have been contacted (NO in step S112), the arithmetic device 11 returns the process to step S111. If all contacts have been contacted (YES in step S112), the arithmetic device 11 ends the process. (Variation 1) In the above-described embodiment, estimation of the user's situation is started based on sound data collected by microphone 17, but estimation of the user's situation may also be started based on location information. In particular, starting estimation of the situation based on location information is effective in situations where the user is unable to speak, such as situations where the user has been abducted, stolen, or wandered away. FIG. 6 is a schematic diagram showing the configuration of a modified example of the emergency notification system according to the present disclosure. An emergency notification system 10d shown in FIG. 6 shows an example of application to child monitoring.

[0039] An emergency notification system 10d is provided in the school bag 21, and the hardware configuration of the emergency notification system 10d is the same as that of the emergency notification system 10a shown in FIG. 2, so detailed description will not be repeated. The emergency notification system 10d may have a hardware configuration that does not include the display 14 or the camera 18. The emergency notification system 10d is a system that monitors children and issues a notification when the distance from the location of the parent (user) carrying the smartphone 100 to the location of the school bag 21 carried by the child exceeds a predetermined distance. The emergency notification system 10d provided in the school bag 21 and the smartphone 100 carried by the parent are capable of communicating with each other.

[0040] The GPS 19 included in the emergency call system 10d is defined as a first position information device, while a GPS (not shown) mounted on a smartphone 100 different from the emergency call system 10d is defined as a second position information device. The calculation device 11 of the emergency call system 10d compares the first position information acquired by the GPS 19 with the second position information acquired by the GPS of the smartphone 100, and if the distance between the first position information and the second position information exceeds a predetermined distance, determines that the child's situation is an emergency situation and sends a report to the smartphone 100. The emergency call system 10d acquires position information from the smartphone 100.

[0041] 6, when the child's situation is an emergency, the smartphone 100 displays on its display the text 101 saying "An abnormality has been detected!", a map 102 showing the child's location 102a, a playback screen 103 for sound data collected by the microphone 17 of the emergency notification system 10d, and a contact police button 104. The playback button 103a is displayed on the sound data playback screen 103, and the word "Police" 104a, which is a contact name, is displayed next to the contact button 104.

[0042] 6, the emergency notification system 10d provided in the school bag 21 is configured to estimate the child's situation based on location information, but the smartphone 100 may be configured to estimate the child's situation based on location information. Furthermore, the emergency notification system 10d may be applied to vehicle theft instead of child monitoring, and the emergency notification system 10d may be provided in the vehicle, and the smartphone 100 may be configured to grasp an emergency situation in which the vehicle has been stolen. (Variation 2) In the above-described embodiment, the user's situation is estimated rule-based based on sound data and location information. However, the user's situation may also be estimated using an estimation model based on machine learning. FIG. 7 is a diagram for explaining the estimation model of the emergency call system according to the present disclosure. The arithmetic unit 11 of the emergency call system 10a shown in FIG. 2 estimates the user's situation based on an estimation model 110 that uses sound data and location information as input data and the user's situation as output data. The estimation model 110 can utilize various machine learning techniques, such as a neural network.

[0043] To generate the estimation model 110, for example, deep learning is used, and input data 120a and correct answer data 120b are used as training data. The training data is a set of input data 120a including known sound data and position information, and correct answer data 120b including known user situations corresponding to the input data 120a.

[0044] Then, when the calculation device 11 executes the training program, it estimates the user's situation from the input data 120a using the estimation model 110 and generates the output data 130. Furthermore, the calculation device 11 compares the supervised data 120b with the calculated output data 130, and performs training to adjust the estimation model 110 so that the loss function between the output data 130 and the supervised data 120b is minimized.

[0045] Therefore, the calculation device 11 can accurately estimate the user's situation from the sound data and the location information by using the trained estimation model 110. As a result, the calculation device 11 can realize a highly accurate emergency notification system by using the trained estimation model 110.

[0046] Although it has been explained that deep learning is used as a machine learning technique to generate the estimation model 110, various machine learning techniques such as linear regression, support vector regression, logistic regression, random forest, support vector machine, and neural network can also be used. (Other variations) In the above-described embodiment, an example has been described in which the emergency notification system 10a simultaneously notifies family members and staff by email and the fire department by telephone, but notifications may be made in stages using multiple communication means. For example, the emergency notification system 10a notifies family members as the primary notification destination by email or telephone, and the family members then notify hospitals and fire departments in stages as necessary. The emergency notification system 10a may also notify link information, etc., to make it easier for family members who receive the notification to notify hospitals and fire departments. In this way, the emergency notification systems 10a to 10c may make notifications in stages using multiple communication means.

[0047] Furthermore, if the emergency notification systems 10a to 10c are unable to communicate with the contact person to whom the notification was made, they may change the communication means and make the notification again. For example, if the emergency notification system 10a attempts to notify a family member by email but is unable to communicate, it may change the communication means to telephone and make the notification again.

[0048] [Aspect] It will be understood by those skilled in the art that the above-described embodiments and their modifications are specific examples of the following aspects.

[0049] (Item 1) An emergency call system according to one aspect is an emergency call system that automatically makes a call depending on a user's situation, a sound collection device that collects the user's voice and sounds around the user; a first location information device that acquires location information; a storage device that stores sound data collected by the sound collection device and first location information acquired by the first location information device; a computing device that estimates a user's situation based on the sound data and the first location information, and selects a plurality of contacts according to the estimated user's situation; The system further includes a communication device that automatically notifies a plurality of contacts selected by the computing device using a preset communication means.

[0050] According to the emergency notification system described in paragraph 1, the user's situation is estimated based on the sound data and the first location information, and multiple contacts are selected according to the estimated user's situation, so that even if the user is unable to recognize the situation, appropriate contacts can be automatically notified, and the emergency notification function is effective.

[0051] (Section 2) The emergency notification system described in Section 1 is implemented in a nursing bed, a school bag, a vehicle, a smart key, a wallet, a writing implement, or a wearable device. This broadly exemplifies devices in which the emergency notification system can be implemented.

[0052] (Item 3) In the emergency notification system described in items 1 or 2, when the computing device detects a sound pressure level equal to or higher than a reference level contained in the sound data, or a sound including a specific keyword registered in advance through sound analysis of the sound data, the computing device starts estimating the user's situation. As a result, the trigger sound starts estimating the user's situation, and an emergency notification can be made automatically.

[0053] (4) The emergency call system according to any one of paragraphs 1 to 3 further includes a second location information device mounted on a device different from the first location information device, The computing device compares the first location information acquired by the first location information device with the second location information acquired by the second location information device, and if the distance between the first location information and the second location information exceeds a predetermined distance, determines that the user's situation is an emergency and automatically reports it using the communication device. This makes it possible to determine that the user's situation is an emergency and automatically report it even if the user is unable to speak.

[0054] (5) In the emergency notification system described in any one of paragraphs 1 to 4, the communication device notifies multiple contacts selected by the arithmetic device using multiple communication means, either stepwise or simultaneously, and if communication with the contact to which the notification was made is not possible, changes the communication means and notifies again. This makes it possible to more reliably notify multiple contacts selected by the arithmetic device.

[0055] (Clause 6) The emergency call system according to any one of clauses 1 to 5 further includes an imaging device that captures an image of the user or the user's surroundings, The imaging device automatically starts capturing images when the computing device starts estimating the user's situation, thereby recording evidence of the emergency situation.

[0056] (7) In the emergency notification system described in any one of paragraphs 1 to 6, the computing device encrypts the sound data collected by the sound collecting device and the image data captured by the image capturing device, and transmits them to an external server or a cloud server via the communication device. This allows evidence of an emergency situation to be securely stored while protecting privacy.

[0057] (8) The emergency call system according to any one of paragraphs 1 to 7, wherein the calculation device estimates the user's situation based on an estimation model having sound data and location information as input data and the user's situation as output data; The estimation model is generated in advance by machine learning based on training data that includes known sound data, location information, and known user situations, allowing for more accurate estimation of the user's situation.

[0058] (Item 9) A program according to one aspect is a program executed by a computing device of an emergency notification system that automatically makes a notification depending on the user's situation, the emergency notification system including a sound collection device that collects a user's voice and sounds around the user, a first position information device that acquires position information of the user, a storage device that stores the sound data collected by the sound collection device and the first position information acquired by the first position information device, a computing device, and a communication device, a process of estimating a situation of the user based on the sound data and the first location information; selecting a plurality of contacts according to the estimated user situation; and automatically notifying the selected plurality of contacts via a communication means of the communication device that is set in advance.

[0059] According to the program described in paragraph 9, the calculation device of the emergency notification system can estimate the user's situation based on the sound data and the first location information, and can execute a process of selecting multiple contacts based on the estimated user's situation.Therefore, even if the user is unable to recognize the situation, the appropriate contacts can be automatically notified, and the emergency notification function will function effectively.

[0060] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the description of the above embodiments, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0061] 10a to 10d Emergency call system, 11 Computing device, 12 Main memory, 13 Input device, 14 Display, 15 Storage device, 16 Communication device, 17 Microphone, 18 Camera, 20 Nursing bed, 21 School bag, 22 Vehicle, 100 Smartphone, 101, 104a Characters, 102 Map, 102a Location, 103 Playback screen 103a Play button, 104 Contact button, 110 Estimation model, 120a Input data, 120b Correct answer data, 130 Output data, 151 OS, 152 Control program, 171 Input processing, 172 A / D conversion processing, 173 Signal preprocessing, 174 Voice analysis processing, 175 Situation estimation processing, 176 Output processing.

Claims

1. An emergency notification system that automatically notifies a user according to the user's situation, a sound collection device that collects the user's voice and sounds around the user; a first location information device that acquires first location information; a storage device that stores sound data collected by the sound collection device; a computing device that estimates a situation of the user based on at least the sound data of the sound data and the first location information, and searches for a plurality of contacts from the storage device or a network according to the estimated situation of the user; and a communication device that automatically notifies the plurality of contacts searched by the computing device using a preset communication means.

2. An emergency notification system as described in claim 1, implemented in a nursing bed, a school bag, a vehicle, a smart key, a wallet, a writing instrument, or a wearable device.

3. The emergency call system according to claim 1 or claim 2, wherein the computing device starts estimating the user's situation when it detects a sound pressure level equal to or higher than a standard contained in the sound data, or when it performs a voice analysis on the sound data and detects a voice containing a specific keyword that has been registered in advance.

4. Further comprising a second position information device mounted on a device different from the first position information device, 3. The emergency notification system according to claim 1, wherein when the user's situation cannot be estimated based on the sound data and the distance from the second location information acquired by the second location information device to the first location information exceeds a predetermined distance, the computing device estimates that the user's situation is an emergency situation and notifies the second location information device via the communication device.

5. 3. The emergency notification system according to claim 1, wherein the communication device notifies the plurality of contacts searched by the computing device using a plurality of communication means in stages or simultaneously, and if communication with the contact to which the notification was made is not possible, the communication means is changed and the notification is made again.

6. further comprising an imaging device that captures an image of the user or the surroundings of the user; The emergency call system according to claim 1 , wherein the imaging device automatically starts imaging when the computing device starts estimating the user's situation.

7. The emergency notification system according to claim 6, wherein the computing device encrypts the sound data collected by the sound collecting device and the image data captured by the image capturing device, and transmits the encrypted data to an external server or a cloud server via the communication device.

8. the computing device estimates the user's situation based on an estimation model having the sound data and the position information as input data and the user's situation as output data; The emergency call system according to claim 1 or 2, wherein the estimation model is generated in advance by machine learning based on training data that is a set of known sound data and location information and known situations of the user.

9. A system including a sound collection device that collects a user's voice and sounds around the user, a first position information device that acquires first position information of the user, a storage device that stores sound data collected by the sound collection device, a calculation device, and a communication device, and a program for automatically reporting in accordance with the user's situation, The computing device: a process of estimating a situation of the user based on at least the sound data of the first position information and the sound data; A process of searching the storage device or the network for a plurality of contacts according to the estimated user situation; and a program for executing a process of automatically notifying the searched contacts via a communication means of the communication device that is preset.

Images