JP7584083B2 - Intercom system, control method, and control program - Google Patents

Description

This disclosure relates to an intercom system, a control method, and a control program.

Conventionally, in apartment complexes and other collective housing, an intercom system is widely known that includes a lobby intercom device installed at the common entrance, multiple intercom master units installed in each dwelling unit, and a control unit that controls these devices. For example, Patent Document 1 discloses an intercom system that receives absence information indicating that a resident is absent, and when a call is made to the intercom master unit of the corresponding dwelling unit, calls are made to an information terminal such as a smartphone.

JP 2020-078036 A

Intercom systems are being introduced not only in apartment complexes and other collective housing, but also in a variety of facilities, including detached houses, offices, factories, medical facilities, and lodging facilities, and high-performance intercom systems with a variety of functions to meet user needs have been proposed. However, user needs for intercom systems are becoming more diverse, and further improvements in usability are expected.

The objective of this disclosure is to provide an intercom system that can further improve usability.

The intercom system according to the present disclosure includes an acquisition unit that acquires sound information within a facility, an alarm unit that outputs predetermined information within the facility, and a control unit that executes processing based on the recognition result of the sound when a trigger word is included in the sound information acquired by the acquisition unit and a sound is acquired following the trigger word, and is characterized in that when the predetermined information is output by the alarm unit, the control unit executes the predetermined processing based on the recognition result of the sound acquired after the output of the predetermined information, even if the sound information does not include the trigger word.

The control method according to the present disclosure is a control method applied to an intercom system configured to execute processing based on the recognition results of sound when sound information includes a trigger word and sound is acquired following the trigger word, and is characterized by including a step of outputting predetermined information, and a step of acquiring sound information after outputting the predetermined information, in which the predetermined processing is executed based on the recognition results of the sound acquired after outputting the predetermined information even if the sound information does not include the trigger word.

The control program according to the present disclosure is a program for implementing the above control method on a computer.

The intercom system according to the present disclosure can further improve usability. The intercom system according to the present disclosure can be used, for example, to watch over children and the elderly, and to check on the safety of residents.

1 is a schematic diagram of an intercom system according to an embodiment; 1 is a block diagram showing a configuration of an intercom master unit according to an embodiment; 4 is a flowchart illustrating a control procedure of the intercom system according to an embodiment. 4 is a flowchart showing a control procedure according to the first embodiment; FIG. 1 is a diagram for explaining a first embodiment. 10 is a flowchart showing a control procedure according to a second embodiment; FIG. 11 is a diagram for explaining a second embodiment.

The embodiments of the present disclosure will be described in detail below with reference to the drawings. The embodiments described below are merely examples, and the present disclosure is not limited to the following embodiments. In addition, selective combinations of multiple embodiments and modified examples described below are included within the scope of the present disclosure.

The subject of the device and system of the present disclosure includes a computer. The computer executes a program to realize the functions of the subject of the device and system of the present disclosure. The computer includes a processor that operates according to the program as the main hardware configuration. The type of processor is not important as long as it can realize the above functions by executing the program. The processor is composed of one or more electronic circuits including an integrated circuit (IC) or a large scale integrated circuit (LSI). The multiple electronic circuits may be integrated into one chip or may be provided on multiple chips. The multiple chips may be integrated into one device or may be provided on multiple devices. The program is stored in a non-transitory storage medium such as a computer-readable ROM, optical disk, or hard disk drive. The program may be stored in the storage medium in advance, or may be supplied to the storage medium via a wide area communication network including the Internet.

Figure 1 is a diagram showing the schematic configuration of an intercom system 1, which is an example of an embodiment. As shown in Figure 1, the intercom system 1 is a system installed in a collective housing complex such as an apartment building, and includes a lobby intercom device 11 provided at a shared entrance 10, multiple intercom master units 21 provided in each dwelling unit 20, and a control device 30 that controls these devices. The control device 30 is a device that provides overall control of the intercom system 1, and is connected to the lobby intercom device 11 and the intercom master unit 21 so that they can communicate with each other.

The intercom system 1 comprises a plurality of entrance handset units 22 provided in each dwelling unit 20. Typically, one intercom master unit 21 and one entrance handset unit 22 are provided in each dwelling unit 20. The intercom master unit 21 is provided inside each dwelling unit 20 and is communicatively connected to one entrance handset unit 22. The entrance handset unit 22 is provided outside each dwelling unit 20 near the entrance. Each dwelling unit 20 may also be provided with a sub-master unit that is communicatively connected to the entrance handset unit 22 via the intercom master unit 21.

The intercom system 1 will be described as being applied to collective housing such as condominiums, dormitories, company housing, and nursing homes, but the application of the intercom system according to the present disclosure is not limited to collective housing. The intercom system according to the present disclosure can be applied to offices, factories, research facilities, medical facilities, nursing care facilities, accommodation facilities, and the like. The intercom system according to the present disclosure may also be applied to detached houses.

The intercom system 1 may include a management office master unit 16 provided in the management room 15. The management office master unit 16 is communicatively connected to the lobby intercom device 11 and the intercom master unit 21 via the control device 30. The management office master unit 16 can transmit various information to the lobby intercom device 11 and the multiple intercom master units 21, for example. In the example shown in FIG. 1, the control device 30 is connected to the server 31 via the gateway 32. The gateway 32 functions as a router. The control device 30 may obtain information necessary for controlling the intercom system 1 from the server 31.

Although the details will be described later, the intercom system 1 has a function of executing a predetermined process based on voice recognition. When sound information containing a trigger word is acquired, the intercom system 1 is configured to execute a process based on the recognition result of the sound acquired following the trigger word. In other words, the intercom system 1 has a "voice assistant function" that recognizes the voice command following the trigger word and executes a process based on the voice command.

The intercom system 1 is also configured to be able to output predetermined information, such as information for confirming the safety of residents. Even if the sound information acquired after outputting the predetermined information does not contain a trigger word, the intercom system 1 executes predetermined processing based on the recognition result of the sound information. Hereinafter, this function may be referred to as a "triggerless voice recognition control function" to distinguish it from a voice assistant function that requires a trigger word.

As described above, the lobby intercom device 11 is installed at the common entrance 10. The lobby intercom device 11 is a device for calling each dwelling unit 20. A visitor can enter the number of the dwelling unit 20 they are visiting into the lobby intercom device 11 to make a call. In the example shown in FIG. 1, a non-contact key detection device 12 that detects a pre-registered card key is installed at the common entrance 10. The card keys are held by the residents of the apartment building, but may be provided to specific visitors such as delivery personnel or newspaper delivery people. Each card key has identification information (ID). Therefore, it is possible to identify the owner of a card key authenticated by the non-contact key detection device 12.

The shared entrance 10 is provided with an entrance door 13 and a locking device 14 that locks the entrance door 13. The locking device 14 is, for example, an auto-locking device that automatically locks the entrance door 13 when the entrance door 13 is closed, and activates an electric lock attached to the entrance door 13 under the control of the control device 30. The entrance door 13 is, for example, an automatic door, and is basically kept locked under normal circumstances.

The lobby intercom device 11 is installed outside the front door 13 of the shared entrance 10. Therefore, in order for a visitor to enter inside the front door 13, the visitor must call the residence 20 they are visiting using the lobby intercom device 11 and ask the resident of that residence 20 to unlock the electric lock on the front door 13. The resident can confirm the visitor using the intercom master unit 21 and operate the intercom master unit 21 to unlock the electric lock. The non-contact key detection device 12, like the lobby intercom device 11, is installed outside the front door 13.

The lobby intercom device 11 typically has a numeric keypad for inputting the dwelling number and a call button for making a call. A visitor can call the dwelling they are visiting by inputting the dwelling number on the numeric keypad and pressing the call button. The lobby intercom device 11 also has, for example, a camera, a microphone, a speaker, a monitor, and a reverse master key insertion section. Residents who own a key can insert the key into the reverse master key insertion section to unlock the electronic lock on the front door 13.

The lobby intercom device 11 captures a visitor with a camera and transmits the image to the intercom master unit 21 that initiated the call. The speaker outputs a ringtone and, if a resident responds, the visitor's voice. The visitor's voice is also picked up by a microphone and transmitted to the intercom master unit 21 of the called dwelling unit 20, allowing a conversation to be made with the resident. The monitor may display, for example, the dwelling unit number entered using the numeric keypad, or the image captured by the camera.

As described above, each dwelling unit 20 is provided with an intercom master unit 21 and an entrance handset 22. The intercom master unit 21 is communicatively connected to the entrance handset 22, and is also communicatively connected to the lobby intercom device 11 via the switch 17 and the control device 30. Using the intercom master unit 21, residents can talk to visitors via the lobby intercom device 11 and the entrance handset 22, and can check video of the visitor. In this embodiment, the intercom master unit 21 is equipped with a voice recognition engine that realizes a voice assistant function.

The intercom master unit 21 is connected to a wide area communication network and may be configured to send a notification to a specific information terminal 60 informing the user that a visitor has arrived. Examples of wide area communication networks include an internet line and a telephone line. The intercom master unit 21 may also be configured to send information such as resident safety confirmation information to the information terminal 60 under triggerless voice recognition control. The information terminal 60 may be any device capable of receiving information from the intercom master unit 21, and examples include a smartphone, a mobile phone, a tablet terminal, and a personal computer.

The front door sub-unit 22 has a call button for calling the intercom master unit 21. The front door sub-unit 22 also has, for example, a camera, a microphone, and a speaker. When the call button is operated, a call is made to the intercom master unit 21 to which the front door sub-unit 22 is connected, and the visitor is photographed by the camera of the front door sub-unit 22, and the video is sent to the intercom master unit 21. The intercom master unit 21 may also have a function for recording video of the visitor.

The control device 30 is communicatively connected to the lobby intercom device 11 at the shared entrance 10, the intercom master unit 21 in each dwelling unit 20, and the management office master unit 16 in the management room 15, and has the function of controlling these devices. The control device 30 is composed of a computer including a processor and memory. The memory is composed of, for example, RAM, ROM, a hard disk, etc. The control device 30 may be composed of one computer or multiple computers. In addition, some or all of the functions of the control device 30 may exist in a device outside the apartment building, such as a server 31.

The configuration of the intercom master unit 21 will be described in detail below with reference to Figure 2. Figure 2 is a block diagram showing the configuration of the intercom master unit 21.

As shown in FIG. 2, the intercom master unit 21 includes a speaker 40, a microphone 41, a call button 42, and an unlock button 43. When the intercom master unit 21 receives a call signal generated based on a call operation of the lobby intercom device 11 or the front door sub unit 22, it outputs a call sound from the speaker 40. The call button 42 is a button for starting and ending a call with the front door sub unit 22, etc. When a resident presses the call button 42 of the intercom master unit 21, he or she can talk to a visitor, etc. via the lobby intercom device 11 or the front door sub unit 22 that has been called. The unlock button 43 is a button for unlocking the electric lock on the front door 13.

The intercom master unit 21 is also provided with a monitor 44. Images captured by the cameras of the lobby intercom device 11 and the entrance slave unit 22 are displayed on the monitor 44. The monitor 44 may also display weather-related information such as emergency earthquake alerts, information on abnormal occurrences such as fires, and messages from the management office 15. An example of the monitor 44 is a touch panel monitor. The intercom master unit 21 is equipped with a communication interface that receives audio and images captured by the entrance slave unit 22, etc.

The intercom master unit 21 further includes a setting operation unit 45 for performing various settings. The intercom master unit 21 has, for example, a flat box-shaped housing, and a call button 42, an unlock button 43, a monitor 44, etc. are provided on the surface of this housing. The setting operation unit 45 may be provided on the surface of the housing, like the call button 42, etc., or may be displayed on a touch panel monitor 44. In this embodiment, the content and activation conditions of the triggerless voice recognition control are set by the setting operation unit 45. Examples of the setting operation unit 45 include multiple buttons, a numeric keypad, a keyboard, etc. displayed on the monitor 44.

The intercom master unit 21 may be provided with buttons for adjusting the call volume, buttons for adjusting the brightness of the monitor 44, buttons for playing recorded video, buttons for configuring the installation settings, and the like. The installation settings may be configured using the setting operation unit 45. The intercom master unit 21 may also be connected to a security system installed in the dwelling unit 20.

The intercom master unit 21 is equipped with a control unit 50 for executing calls with the lobby intercom device 11 and the entrance slave unit 22, displaying images captured by the camera, recording, etc. The control unit 50 is composed of a microcomputer having a processor 51 and a memory 52, and is built into the intercom master unit 21. The memory 52 is composed of, for example, a RAM, a ROM, a hard disk, etc., and stores a control program and various setting information. The processor 51 realizes the functions of each processing unit of the control unit 50 by reading and executing the control program installed in the memory 52.

The control unit 50 includes a first processing unit 53 that, when a call signal is received from the lobby intercom device 11 or the front door unit 22, outputs a call sound from the speaker 40 and executes processing to display an image of the visitor on the monitor 44. When the talk button 42 is pressed, the first processing unit 53 also makes the lobby intercom device 11 or the front door unit 22, which are the source of the call signal, available for communication, and transmits the audio signal acquired by the microphone 41 to the device in communication. The first processing unit 53 controls the operation of the intercom master unit 21 as a general intercom system.

The control unit 50 further includes a second processing unit 54, a third processing unit 55, and a fourth processing unit 56 as processing units for executing the voice assistant function and the triggerless voice recognition control function. In this embodiment, all of the components for executing the voice assistant function and the like are present in the intercom master unit 21. Specifically, the microphone 41 of the intercom master unit 21 serves as an acquisition unit that acquires sound information, and the speaker 40 serves as an alarm unit that outputs predetermined information such as safety confirmation information.

When a trigger word is included in the sound information acquired by the microphone 41 and a sound is acquired following the trigger word, the second processing unit 54 executes processing based on the recognition result of the sound. That is, the second processing unit 54 recognizes the trigger word and voice command from the sound information acquired by the microphone 41, generates a recognition result, and executes a predetermined process based on the result. The intercom master unit 21 is equipped with a voice recognition engine that recognizes the trigger word and voice command from the acquired sound information.

The trigger word is a keyword used to activate the voice assistant function, and is also commonly called a wake word. If the trigger word is not recognized from the sound information, the voice assistant function will not be executed. Examples of trigger words include the name of the intercom system 1, the name of the software for the voice assistant function (or voice recognition engine), and words pre-registered in the system. In addition, the trigger word may be arbitrarily set by the user by operating the setting operation unit 45.

The scope of application of the voice assistant function may cover all operations that can be performed by operating the intercom master unit 21. Specific examples include starting and ending calls with the lobby intercom device 11 and the front door unit 22, unlocking the electric lock on the front door 13, recording and playing back video, and adjusting the call volume. For example, a user can start a call with the front door unit 22 by uttering a trigger word toward the intercom master unit 21 and then uttering a voice command such as "put on call". The scope of application of the voice assistant function may be limited to specific operations, such as operations in a non-call state.

As described above, the control unit 50 executes triggerless voice recognition control that does not require a trigger word. The control unit 50 first outputs predetermined information from the speaker 40. Then, when the predetermined information is output, the control unit 50 executes predetermined processing based on the sound recognition result acquired after the output of the predetermined information, even if the sound information acquired by the microphone 41 does not include a trigger word. In this embodiment, the predetermined information is output by the function of the third processing unit 55, and processing based on the sound recognition result is executed by the function of the second processing unit 54.

When a predetermined activation condition is met, the third processing unit 55 causes the speaker 40 of the intercom master unit 21 to output predetermined information. The output of this predetermined information activates triggerless voice recognition control. In this case, the second processing unit 54 executes a predetermined process based on the sound recognition results obtained after the output of the predetermined information, even if the trigger word is not recognized by the voice recognition engine. In other words, in this control mode, the output of the predetermined information by the third processing unit 55 serves as a trigger. Then, the user only needs to utter a command word, and the predetermined process is automatically executed.

The third processing unit 55 may output the specified information multiple times if the voice recognition engine does not generate an appropriate recognition result for executing the specified process. An upper limit may be set for the number of times the specified information is output. Alternatively, an upper limit may be set for the time for which the specified information is output. That is, the third processing unit 55 outputs the specified information until, for example, the number or time of recognition errors of a voice command by the voice recognition engine exceeds a predetermined threshold. The second processing unit 54 may execute a process different from the process based on the appropriate authentication result if an appropriate recognition result is not generated.

Examples of activation conditions for outputting specific information include authentication of a specific card key by the non-contact key detection device 12 at the shared entrance 10, the arrival of a preset time, and acquisition of specific information such as an emergency earthquake alert. Since each card key has identification information, for example, by registering the identification information of a card key held by a child in advance as information for activating triggerless voice recognition control, the authentication information of the specific card key can be used as the activation condition.

The third processing unit 55, for example, outputs predetermined information from the speaker 40 of the intercom master unit 21 after a predetermined time has elapsed after a specific card key has been authenticated by the non-contact key detection device 12. At this time, the predetermined information may be output from the speaker of the lobby intercom device 11 or the entrance sub unit 22. The third processing unit 55 may also output the predetermined information from the speaker 40 when the time reaches a preset time or when specific information such as an emergency earthquake alert is obtained.

Examples of the above-mentioned predetermined information include information for confirming the safety of residents and information for managing the activities of residents. Information for confirming safety is output, for example, to confirm whether children have returned home, the health of elderly people, or the condition of residents when an abnormality such as an earthquake occurs. Information for managing activities is output, for example, to confirm whether medication has been taken, health checks have been performed, rehabilitation exercise has been performed, medical equipment has been inspected, or residents have visited a hospital, and in some cases to encourage these actions. In addition, various information to be provided to residents, such as news, weather information, neighborhood information, and announcements about the apartment building, may be output as predetermined information.

The above-mentioned specified information may be displayed on the monitor 44, but in this case, it is preferable to output a sound from the speaker 40 to notify that the information has been displayed on the monitor 44. It is more preferable that the sound output from the speaker 40 includes voice. The voice may be a synthesized voice preregistered in the intercom master unit 21, or may be a voice recorded by a resident or the like. Furthermore, if the specified information is information for confirming safety or information for activity management, this voice includes an inquiry to the resident. Furthermore, the voice including the inquiry may include guidance on how to respond to the inquiry.

Examples of voice inquiries include those requiring a response such as "It is now 8 o'clock. Have you taken your blood pressure?" or "What was your blood pressure?". Examples of guidance on how to respond to inquiries include those specifying the response method such as "It is now 8 o'clock. Have you taken your blood pressure? Please answer with yes/no" or "What was your blood pressure? Please answer with your systolic blood pressure and then your diastolic blood pressure." Outputting guidance on how to respond makes it easier for residents to respond and also improves the accuracy of voice recognition.

In this embodiment, when a voice including an inquiry is output from the speaker 40, the resident's response (sound information) following the voice is acquired by the microphone 41, and a recognition result is generated by the voice recognition engine. The voice recognition engine converts the sound information acquired by the microphone 41 into text, for example, and recognizes commands for executing a specified process. This series of processes is executed by the function of the second processing unit 54. The voice recognition engine may be started after the inquiry voice is output, or an electronic sound such as "beep" may be output from the speaker 40 to notify the start of the engine.

After outputting the voice from the speaker 40, the control unit 50 may record the sound information collected by the microphone 41 regardless of the recognition result of the voice recognition engine. This sound information may be recorded in the memory 52 of the intercom master unit 21, or may be transmitted to at least one of the management office master unit 16, the server 31, and the information terminal 60 and recorded by these devices. This recording function allows, for example, a resident to check the content of the response. It also allows the manager or the like to check the sound information.

The above-mentioned predetermined process may, for example, when safety confirmation information is output, output of safety information that can be received by the information terminal 60 using information related to the information terminal 60 that has been registered in advance. The intercom master unit 21 is connected to a wide area communication network such as the Internet or a telephone line, and therefore can transmit various information to the information terminal 60 such as a smartphone via the wide area communication network. For example, the memory 52 of the intercom master unit 21 stores information related to the information terminal 60, such as an email address, a telephone number, and account information for a specified app.

The second processing unit 54, for example, sends an email containing safety information to a pre-registered email address. The safety information contained in the email varies depending on the resident's response, i.e., the recognition result generated by the voice recognition engine, and whether or not a recognition result has been generated. In other words, the specified processing to be executed varies depending on the presence or absence of a recognition result and its content. Note that if a recognition result is not generated, for example, no processing is executed depending on the content of the specified information to be output.

For ease of explanation, the predetermined process executed when a recognition result is generated based on a positive response to predetermined information including an inquiry, indicating that the person has returned home, is in good health, is safe, etc., is referred to as the "first process." On the other hand, the predetermined process executed when a recognition result is generated based on a negative response, or when no recognition result is generated, is referred to as the "second process."

After predetermined information is output from the speaker 40, the second processing unit 54 ends the triggerless voice recognition control when a predetermined stop condition is met. That is, when the stop condition is met, the second processing unit 54 does not execute processing based on sound information that does not contain a trigger word. When the stop condition is met, a trigger word is required to operate the intercom system 1 by voice until the above-mentioned activation condition is met again. By setting a stop condition and automatically terminating the triggerless voice recognition control, malfunction of the system can be effectively suppressed.

The above stop conditions include, for example, that the time elapsed since the specified information was output exceeds a threshold, or that the number of times the specified information is output exceeds a threshold. In general, such a stop condition is met when appropriate voice recognition for executing the first process cannot be performed. The thresholds for the elapsed time and the number of times output are not particularly limited, and may be arbitrarily set by operating the setting operation unit 45. The stop conditions may also include whether or not sound information has been recognized, or whether or not processing based on sound information has been performed. When appropriate voice recognition for executing the first process has been performed, such a stop condition is met.

The intercom system 1 may be provided with at least one of a setting means for setting the content of the above-mentioned specified information and a setting means for setting the above-mentioned activation conditions. In this embodiment, the specified information and activation conditions can be set by operating the setting operation unit 45. The user can also set the content of the specified processing to be executed based on the voice recognition result and the information required for the specified processing. Note that the intercom system 1 may have several pieces of specified information, activation conditions, and specified processing preset. Triggerless voice recognition control may be executed based on the content selected by the user from among them.

The fourth processing unit 56 executes a process for setting the contents of the specified information to be output from the speaker 40. For example, when the setting operation unit 45 is operated, the fourth processing unit 56 stores in the memory 52 the specified information and the contents of the specified process input by the operation. The fourth processing unit 56 also executes a process for setting the start-up conditions. For example, when the setting operation unit 45 is operated, the fourth processing unit 56 stores in the memory 52 the contents of the start-up conditions input by the operation. The first processing unit 53 and the second processing unit 54 read out this setting information from the memory 52 and execute triggerless voice recognition control.

The control procedure (operation) of the intercom system 1 will be described below with reference to Figure 3. Figure 3 is a flowchart showing an example of triggerless voice recognition control.

Triggerless voice recognition control, in summary, is a control that includes a first step of outputting predetermined information, and a second step of acquiring sound information after the predetermined information is output. In the second step, a predetermined process is executed based on the sound recognition result acquired after the predetermined information is output, even if the sound information does not include a trigger word. A control program for implementing triggerless voice recognition control by a computer is stored in memory 52 of intercom master unit 21.

As shown in FIG. 3, when a predetermined activation condition is met, triggerless voice recognition control is started (S1). In S1, it is determined whether the activation condition for triggerless voice recognition control is met. The activation condition is met, for example, when a pre-specified target person is detected at the shared entrance 10 or the like. The method of detecting the target person is not particularly limited, and examples include a method of detection from authentication information of a card key by the non-contact key detection device 12, and a method of detection from face authentication information of a camera installed at the shared entrance 10 or the like. The activation condition may also be met when a pre-set time is reached or when specific information such as an emergency earthquake alert is acquired.

If it is determined in S1 that the activation conditions are met, predetermined information such as information for confirming safety and information for managing activities is output (S2). The predetermined information may be video or text information output to the monitor 44 of the intercom master unit 21, but is preferably output as audio information from the speaker 40. In this embodiment, the predetermined information is output from the intercom master unit 21 to the resident in the dwelling unit 20, but the predetermined information may also be output from the lobby intercom device 11 or the entrance unit 22.

When the specified information is output in S2, a voice command is recognized from the sound information acquired thereafter, and a recognition result is generated (S3). The voice command in S3 is an instruction for executing the process in S4. A voice recognition engine installed in the intercom master unit 21 converts the acquired sound information into text and recognizes the voice command (generation of recognition result). Sound information is acquired by the microphone 41 of the intercom master unit 21, but if the sound information cannot be acquired or if the appropriate voice command cannot be recognized, no recognition result is generated. In that case, the process may return to S2 and output the specified information again.

In S3, even if the sound information acquired by the microphone 41 does not contain a trigger word, a sound information recognition result is generated. In contrast, in the case of control by a voice assistant function in which S2 does not exist, a sound information recognition result is generated only if the sound information acquired by the microphone 41 contains a trigger word. In other words, in control by the voice assistant function, if the trigger word is not recognized, a sound information recognition result is not generated. In S3, a sound information recognition result is generated regardless of whether or not a trigger word is recognized.

If an appropriate recognition result is generated in S3, a first process is executed based on the recognition result (S4). An example of the first process in S4 is to send a notification to a pre-registered email address or the like informing the user that the resident is safe when safety confirmation information is output in S2. On the other hand, if an appropriate recognition result is not generated in S3, a second process different from the process in S4 is executed (S5). An example of the second process in S5 is to send a notification to a pre-registered email address or the like informing the user that there is no response from the resident.

When the second process is executed, it may be possible to call the intercom master unit 21 from the information terminal 60. The intercom system 1 may be configured, for example, to call the intercom master unit 21 from the information terminal 60 and enable a call with the intercom master unit 21 even if the call button 42 is not pressed. The second process may also be a process that calls the information terminal 60 from the intercom master unit 21 and enables a call with the intercom master unit 21. Such a call function may be activated only when the second process is executed. Software for realizing this call/call function may be installed in the intercom master unit 21 and the information terminal 60.

A resident outside the dwelling unit 20 can, for example, call the intercom master unit 21 from the information terminal 60, or respond to a call from the intercom master unit 21 to have a sound output from the speaker 40 of the intercom master unit 21 into the room of the dwelling unit 20. In addition, if a security system is connected to the intercom master unit 21, the resident may be able to access the security system from the information terminal 60 via the intercom master unit 21 and check the situation inside the room of the dwelling unit 20.

After the specified information is output in S2, triggerless voice recognition control is terminated when a specified stop condition is met (S6). When the stop condition is met, a trigger word is required for subsequent voice control of the intercom system 1. The stop condition is met, for example, when the elapsed time since the specified information was output exceeds a threshold value, or when the number of times the specified information has been output exceeds a threshold value. Alternatively, the stop condition is met when sound information is recognized, or when control based on sound information is executed.

When the process proceeds from S3 to S5, triggerless voice recognition control may be continued until a manual stop operation is performed. The manual stop operation can be performed, for example, by operating the intercom master unit 21.

Below, a specific example of triggerless voice recognition control will be described with reference to Figures 4 to 7. Figures 4 and 5 show an example of triggerless voice recognition control for confirming that a child has returned home.

In the example shown in Figures 4 and 5, when a child's return home is detected at the shared entrance 10, triggerless voice recognition control is started (S11). Specifically, a child's return home is detected at the shared entrance 10 by the non-contact key detection device 12 authenticating the card key held by the child, or by recognizing the child's face from video captured by the camera of the lobby intercom device 11 or a surveillance camera installed at the shared entrance 10. The intercom system 1 starts triggerless voice recognition control, for example, when a process to unlock the electric lock on the entrance door 13 is performed and it is determined that the process was initiated by a child.

When a child's return home is detected at the shared entrance 10 based on authentication information of the card key, etc., safety (return home) confirmation information is output after a predetermined time has elapsed since the detection to confirm that the child has arrived safely inside the dwelling unit 20 (S12, S13). The return home confirmation information includes a voice inquiry requiring a response, such as "Please respond when XX-chan arrives home," and is output from the speaker 40 of the intercom master unit 21. The predetermined time in S12 is determined, for example, taking into consideration the time it takes to arrive from the shared entrance 10 to the dwelling unit 20.

The inquiry to confirm that the child has returned home may be output multiple times at predetermined intervals after the above-mentioned predetermined time has elapsed. The voice of the inquiry may be a synthesized voice, or may be a voice recorded by the resident. The voice of the inquiry may also include guidance on how to respond, such as "When XX-chan gets home, please respond with 'I'm home.'" Alternatively, when the child's return is detected at the shared entrance 10, or after the above-mentioned predetermined time has elapsed, a notification of the child's return may be sent from the intercom system 1 to the information terminal 60, so that the information terminal 60 may be configured to make an inquiry to confirm that the child has returned home in real time.

When a voice inquiry about returning home is output from the speaker 40 of the intercom master unit 21, the voice command is recognized from the sound information acquired by the microphone 41 and a recognition result is generated (S14). In S14, even if the acquired sound information does not contain a trigger word, a recognition result for the sound information is generated. In other words, if a child who has returned home hears the voice inquiry about returning home output from the speaker 40 and responds by saying "I'm home," indicating that he or she has returned home, the response is recognized as a voice command for executing the process described below, even if it does not contain a trigger word.

When a response indicating that the user has returned home is made and voice recognition is performed, the device outputs return-home confirmation information (S15). Specifically, the device uses information related to the information terminal 60, such as an email address, telephone number, and account information for a specific application, stored in the memory 52 of the intercom master unit 21, to output a notification that can be received by the information terminal 60, such as sending an email containing the return-home confirmation information to the email address. The intercom master unit 21 may record a response to the inquiry regarding return-home confirmation and output this as the return-home confirmation information. When a response indicating that the user has returned home is recognized or when the return-home confirmation information is output, the device terminates the triggerless voice recognition control, assuming that a specific stop condition has been met.

On the other hand, if a response indicating the return of the user is not recognized, the device outputs information indicating the return of the user not being confirmed, such as by sending an email containing such information to a registered email address (S16). In this case, the information terminal 60 may call the intercom master unit 21, enabling a call to be made with the intercom master unit 21. Alternatively, in S16, instead of or in addition to outputting the information indicating the return of the user not being confirmed, the intercom master unit 21 may call the information terminal 60, enabling a call to be made with the intercom master unit 21, and access to the security system may be permitted from the information terminal 60.

In this case, a resident outside the dwelling unit 20 can have the speaker 40 of the intercom master unit 21 output a voice message encouraging a response from inside the dwelling unit 20. In addition, the resident can access the security system from the information terminal 60 and check the state of the interior of the dwelling unit 20 using the security system's camera. When a response indicating the return of the resident is recognized from the output of the voice message encouraging a response, or when return of the resident confirmation information is output, it is determined that a specified stop condition has been met and the triggerless voice recognition control is terminated. On the other hand, if a response indicating the return of the resident is still not recognized, the triggerless voice recognition control may be continued until a manual stop operation is performed.

Figures 6 and 7 show an example of triggerless voice recognition control for confirming medication taking.

In the examples shown in Figures 6 and 7, triggerless voice recognition control is started when the current time becomes a preset time (S21). The intercom master unit 21 has information on the current time and the preset time, and when the two times match, it outputs activity management information for confirming whether or not the medication has been taken, such as "Have you taken your medicine?" (S22). This activity management information is a voice inquiry that requires a response as to whether or not the medication has been taken, and is output from the speaker 40 of the intercom master unit 21. The voice inquiry may include guidance on how to respond, such as "Have you taken your medicine? Please answer yes or no."

The inquiry about taking medicine may be output multiple times at predetermined intervals until a voice response is recognized. In addition, when the current time reaches a preset time, information on the time of taking medicine may be transmitted from the intercom system 1 to the information terminal 60, and the information terminal 60 may be configured to make an inquiry in real time. In addition to or instead of the inquiry about taking medicine, inquiries about health conditions including whether or not the resident has received a drug injection or performed rehabilitation exercises, blood pressure, blood glucose measurement results, etc. may be made according to the resident's needs. Such settings may be made by the resident or by a care worker such as a care manager.

When a voice inquiry about whether or not the medicine has been taken is output from the speaker 40 of the intercom master unit 21, a voice command is recognized from the sound information acquired by the microphone 41, and a recognition result is generated (S23). In S23, even if the acquired sound information does not contain a trigger word, a recognition result for the sound information is generated. For example, if an elderly person who should take medicine hears the voice inquiry about whether or not the medicine has been taken output from the speaker 40 and responds by saying, "I took it," indicating that the medicine has been taken, the response is recognized as a voice command for executing the process described below, even if it does not contain a trigger word.

When a response indicating that the medicine has been taken is given and voice recognition is performed, the taking confirmation information is output (S24). At this time, a list (taking history) summarizing the number of times the medicine has been taken, the taking time, etc. may be created and stored in memory 52, etc., and this list may be output as the taking confirmation information. The intercom master unit 21 outputs a notification that can be received by the information terminal 60, for example, by sending an email including the taking confirmation information to a registered email address. When a response indicating that the medicine has been taken is recognized or when the taking confirmation information is output, it is considered that a predetermined stop condition has been met, and the triggerless voice recognition control is terminated.

On the other hand, if a response indicating that the medication has been taken is not recognized, unconfirmed medication information is output, such as by sending an email containing unconfirmed medication information to the registered email address (S25). If a response indicating that the medication has not been taken is made and voice recognition is performed in S23, such as "I haven't taken it yet," unconfirmed medication information is output, and a voice message encouraging the user to take the medication, such as "Please take your medication as soon as possible," may be output from the speaker 40 of the intercom master unit 21.

If the voice recognition engine does not recognize a response, the sound information collected by the microphone 41 at that time may be acquired. For example, the intercom master unit 21 outputs an inquiry about whether the medication has been taken, and then records the sound information collected by the microphone 41. The intercom master unit 21 may also transmit the recorded sound information to at least one of the management office master unit 16, the server 31, and the information terminal 60. In particular, in the case of elderly people, it is expected that the content of the speech will not be recognized by the voice recognition engine, but if the sound information is recorded, the manager, etc., can listen to the sound information and check whether the medication has been taken, the resident's situation, etc.

Furthermore, if there is no response in S23 and no voice recognition result is generated, unconfirmed medication information may be output, and the information terminal 60 may call the intercom master unit 21, enabling a conversation with the intercom master unit 21. Alternatively, in S25, instead of or in addition to outputting the unconfirmed medication information, the intercom master unit 21 may call the information terminal 60, enabling a conversation with the intercom master unit 21, or access to the security system may be permitted from the information terminal 60. If there is no response, it is assumed that an abnormality has occurred in the resident's health, so by enabling such a response, the abnormality with the resident can be identified quickly.

As described above, with the intercom system 1, various operations can be performed on the intercom master unit 21 by issuing a voice command following a trigger word. Furthermore, by outputting voice inquiries such as safety confirmation information and activity management information from the speaker 40 of the intercom master unit 21, it is possible to, for example, confirm that children have returned home, manage the health of elderly people, and confirm the safety of residents in the event of an earthquake, etc.

When a resident hears the voice inquiry, they simply respond in accordance with the inquiry, and the intercom system 1 automatically executes the specified process. There is no need to speak a trigger word in this case. In other words, by using the nearby intercom system 1, it is possible to quickly watch over children or the elderly, check on the safety of residents, and so on, with a simple voice response that does not require a trigger word.

In addition, in the intercom system 1, the contents of the triggerless voice recognition control, including the specified information to be output from the speaker 40, the activation conditions for outputting the specified information, and the specified processing to be executed in response to the resident's response, can be arbitrarily set according to the needs of each user. Residents, etc. can easily set and change these contents by operating the setting operation unit 45.

Intercom system 1 can be an extremely useful, convenient, and user-friendly system in today's world where user needs and lifestyles are becoming more diverse.

The above embodiment may be modified as appropriate without impairing the purpose of this disclosure. For example, in the above embodiment, the voice recognition engine is mounted on the intercom master unit 21, but the voice recognition engine may be mounted on the control device 30, the lobby intercom unit 11, the entrance unit 22, the server 31, or another device connected to the intercom system 1 so as to be able to communicate with it.

In addition, in the above-mentioned triggerless voice recognition control, a voice inquiry that requests a response is output as the specified information, but information that does not request a response may also be output. Examples of information that does not require a response include various types of information, such as news of a genre preset by the resident, information on new product releases, sales information, weather information, the day's schedule, and apartment announcements. When such information is output, the desired process can be executed without a trigger word by issuing a voice command that requests the transmission of information, such as "email that information," or a voice command that requests additional information, such as "what's the weather tomorrow?".

1 Intercom system, 10 Shared entrance, 11 Lobby intercom device, 12 Non-contact key detection device, 13 Entrance door, 14 Locking device, 15 Management room, 16 Management office master unit, 17 Splitter, 20 Residence, 21 Intercom master unit, 22 Entrance slave unit, 30 Control device, 31 Server, 32 Gateway, 40 Speaker, 41 Microphone, 42 Talk button, 43 Unlock button, 44 Monitor, 45 Setting operation unit, 50 Control unit, 51 Processor, 52 Memory, 53 First processing unit, 54 Second processing unit, 55 Third processing unit, 56 Fourth processing unit, 60 Information terminal

Claims (8)

An acquisition unit that acquires sound information within a facility;
a notification unit that outputs predetermined information within the facility;
a control unit that, when a trigger word is included in the sound information acquired by the acquisition unit and a sound is acquired following the trigger word, executes a process based on a recognition result of the sound;
Equipped with
The predetermined information includes a voice of an inquiry to a person in the facility and guidance on how to respond to the inquiry,
An intercom system in which, when the specified information is output by the alarm unit, the control unit executes a specified processing based on the sound recognition result obtained after the output of the specified information, even if the sound information does not include the trigger word. 2. The intercom system according to claim 1 , further comprising a first setting means for setting the content of said predetermined information. a second setting means for setting a start condition for outputting the predetermined information;
The intercom system according to claim 1 , wherein the notification unit outputs the predetermined information when the activation condition is satisfied. The intercom system according to any one of claims 1 to 3, wherein the control unit does not execute processing based on the sound information that does not include the trigger word when a predetermined stop condition is met after the predetermined information is output by the alarm unit. The intercom system according to claim 4 , wherein the stop condition includes a condition in which an elapsed time since the predetermined information was output exceeds a threshold, or a condition in which the number of times the predetermined information is output exceeds a threshold. The intercom system according to claim 4 or 5 , wherein the stop condition includes that the sound information has been recognized or that a process based on the sound information has been executed. A control method applied to an intercom system configured to execute a process based on a recognition result of a sound when a trigger word is included in sound information and a sound is acquired following the trigger word, comprising:
outputting predetermined information including a query to persons within the facility and instructions on how to respond to said query ;
a step of acquiring the sound information after the predetermined information is output, and a step of executing a predetermined process based on a sound recognition result acquired after the predetermined information is output even if the sound information does not include the trigger word;
A control method comprising: A control program for implementing the control method according to claim 7 on a computer.

Images