Information terminal, intercom system, voice recognition system, control method, and program

Abstract

To provide an information terminal capable of improving the function of voice recognition.SOLUTION: An information terminal 10 includes a voice recognition unit 182, a control processing unit 183, and a communications unit (first communications unit 11). The voice recognition unit 182 performs voice recognition based on voices input into microphone 131. The control processing unit 183 performs the control based on the voice recognition result by the voice recognition unit 182. The communications unit performs communication including calling with other devices. The voice recognition unit 182 performs voice recognition using a first method, or a second method, which is simpler than the first method. The voice recognition unit 182 performs voice recognition using the first method in standby mode and performs voice recognition using the second method in call mode with another device.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure generally relates to an information terminal, an intercom system, a voice recognition system, a control method, and a program. More specifically, the present disclosure relates to an information terminal having a communication function including a call with another terminal, an intercom system including the information terminal, a voice recognition system that communicates with the information terminal, a control method for the information terminal, a control method for the voice recognition system, and a program.

Background Art

[0002] Patent Document 1 discloses an intercom system used in apartment buildings and the like. The intercom system of Patent Document 1 includes a lobby intercom installed at a common entrance of an apartment building, an intercom master unit installed in each dwelling unit, and a doorphone slave unit installed outside (at the entrance of) each dwelling unit.

[0003] In such an intercom system, a visitor calls the intercom master unit using the lobby intercom. In response to this call, when a resident of the dwelling unit performs a predetermined operation on the intercom master unit, a call is started between the lobby intercom and the intercom master unit. Similarly, when calling the intercom master unit from the doorphone slave unit, when a resident of the dwelling unit performs a predetermined operation on the intercom master unit in response to the call, a call is started between the doorphone slave unit and the intercom master unit.

Prior Art Literature

Patent Literature

[0004]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0005] Incidentally, there are systems that perform voice recognition based on the voices spoken by residents of a dwelling unit as part of a prescribed operation. When using voice recognition in information terminals (e.g., intercom master units) or intercom systems, improvements to the voice recognition functionality are desired.

[0006] This disclosure is made in view of the above circumstances and aims to provide an information terminal, an intercom system, a voice recognition system, a control method, and a program that can improve the functionality of voice recognition. [Means for solving the problem]

[0007] An information terminal according to one embodiment of the present disclosure comprises a speech recognition unit, a control processing unit, and a communication unit. The speech recognition unit performs speech recognition based on speech input to a microphone. The control processing unit performs control based on the speech recognition result of the speech recognition unit. The communication unit performs communication, including calls with other terminals. The speech recognition unit performs speech recognition using a first method or a second method that is simpler than the first method. The speech recognition unit performs speech recognition using the first method in a standby state and uses the second method when in a call state with other terminals. The first method is a method in which the control processing unit performs first command voice recognition for a first control command relating to a predetermined control performed in the standby state based on a first dictionary. The second method is a method in which the control processing unit performs second command voice recognition for a second control command relating to a specific control performed in the call state based on a second dictionary. The number of first control commands registered in the first dictionary is greater than the number of second control commands registered in the second dictionary.

[0008] An intercom system according to one aspect of this disclosure comprises the above-mentioned information terminal and the other terminal.

[0009] An information terminal according to one embodiment of the present disclosure comprises a speech recognition unit, a control processing unit, a first communication unit, and a second communication unit. The speech recognition unit performs speech recognition based on speech input to a microphone. The first communication unit performs communication, including calls with other terminals. The second communication unit communicates with a speech recognition system that performs command speech recognition for control commands related to the control performed by the control processing unit. The information terminal has a first mode and a second mode. In the first mode, the speech recognition unit performs trigger speech recognition to determine whether or not a trigger word is included in the speech input to the microphone. In the first mode, if the trigger word is included, the information terminal communicates with the speech recognition system to perform command speech recognition. In the second mode, the speech recognition unit does not perform trigger speech recognition, but communicates with the speech recognition system to perform command speech recognition. The information terminal operates in the first mode in a standby state and in the second mode when in a call state with other terminals.

[0010] A speech recognition system according to one aspect of the present disclosure comprises a communication unit that communicates with an information terminal and a speech recognition unit. When the speech recognition unit receives voice input to a microphone from the information terminal via the communication unit, it performs speech recognition on the voice. The speech recognition unit has a first mode and a second mode. In the first mode, it performs trigger speech recognition to determine whether or not a trigger word is included in the voice input to the microphone. In the first mode, if the trigger word is included, it performs command speech recognition for control commands related to control performed by the control processing unit of the information terminal. In the second mode, it performs command speech recognition without performing trigger speech recognition. The speech recognition system operates in the first mode when the information terminal is in a standby state and operates in the second mode when the information terminal is in a call state with another terminal.

[0011] One embodiment of the present disclosure is a control method for controlling an information terminal. The control method includes a speech recognition step, a control processing step, and a communication step. In the speech recognition step, speech recognition processing is performed based on speech input to a microphone. In the control processing step, control is performed based on the speech recognition result of the speech recognition step. In the communication step, communication is performed, including a call with another terminal. The speech recognition processing is performed using a first method or a second method that is simpler than the first method of speech recognition. In the standby state, speech recognition is performed using the first method, and in the call state with the other terminal, speech recognition is performed using the second method. The first method is a method in which, in the control processing step, first command speech recognition for a first control command relating to a predetermined control performed in the standby state is performed based on a first dictionary. The second method is a method in which, in the control processing step, second command speech recognition for a second control command relating to a specific control performed in the call state is performed based on a second dictionary. The number of first control commands registered in the first dictionary is greater than the number of second control commands registered in the second dictionary.

[0012] A control method according to one aspect of the present disclosure is a control method for controlling an information terminal. The control method includes a speech recognition step, a control processing step, a first communication step, and a second communication step. In the speech recognition step, speech recognition processing is performed based on speech input to a microphone. In the first communication step, communication is performed, including a call with another terminal. In the second communication step, communication is performed with a speech recognition system that performs command speech recognition for control commands related to the control performed in the control processing step. The information terminal has a first mode and a second mode. In the first mode, trigger speech recognition is performed in the speech recognition step to determine whether or not a trigger word is included in the speech input to the microphone. In the first mode, if the trigger word is included, communication is performed with the speech recognition system in the second communication step to cause the system to perform command speech recognition. In the second mode, trigger speech recognition is not performed in the speech recognition step, and communication is performed with the speech recognition system in the second communication step to cause the system to perform command speech recognition. In the control method described above, the information terminal is controlled to operate in the first mode when in standby mode and in the second mode when in call mode with the other terminal.

[0013] One embodiment of the present disclosure is a control method for controlling a speech recognition system. The control method includes a communication step of communicating with an information terminal and a speech recognition step. In the speech recognition step, when speech input to a microphone is received from the information terminal in the communication step, speech recognition processing is performed on the speech. The speech recognition system has a first mode and a second mode. In the first mode, in the speech recognition step, trigger speech recognition is performed to determine whether or not a trigger word is included in the speech input to the microphone. In the first mode, if the trigger word is included, command speech recognition is performed for control commands related to control performed by the control processing unit of the information terminal. In the second mode, in the speech recognition step, command speech recognition is performed without performing trigger speech recognition. The control method operates in the first mode when the information terminal is in a standby state and operates in the second mode when the information terminal is in a call state with another terminal.

[0014] A program according to one aspect of this disclosure is a program that causes one or more processors to execute any of the control methods described above. [Effects of the Invention]

[0015] According to this disclosure, there is an advantage in that it is possible to improve the functionality of speech recognition. [Brief explanation of the drawing]

[0016] [Figure 1] Figure 1A is a block diagram illustrating the configuration of an information terminal according to one embodiment. Figure 1B is a block diagram illustrating the configuration of the operating section of the same information terminal. [Figure 2] Figure 2 is a block diagram illustrating the configuration of an intercom system equipped with the same information terminal. [Figure 3] Figure 3 is a flowchart illustrating an example of call-related operations (Operation Example 1) on the same information terminal. [Figure 4]FIG. 4 is a conceptual diagram for explaining the above-mentioned operation example (Operation Example 1). [Figure 5] FIG. 5A and FIG. 5B are conceptual diagrams for explaining an operation example (Operation Example 2) relating to a predetermined operation (operation of a security alarm set) in the above-mentioned information terminal. [Figure 6] FIG. 6 is a conceptual diagram for explaining an operation example (Operation Example 3) relating to a predetermined operation (operation of door monitor confirmation) in the above-mentioned information terminal. [Figure 7] FIG. 7 is a conceptual diagram for explaining an operation example (Operation Example 4) relating to a predetermined operation (operation of recorded video playback) in the above-mentioned information terminal. [Figure 8] FIG. 8 is a conceptual diagram for explaining an operation example (Operation Example 5) relating to a predetermined operation (operation of air conditioner control) in the above-mentioned information terminal. [Figure 9] FIG. 9 is a conceptual diagram for explaining an operation example (Operation Example 6) relating to a predetermined operation (operation of scene control) in the above-mentioned information terminal. [Figure 10] FIG. 10 is a block configuration diagram for explaining the configuration of first modification of an intercom system including the above-mentioned information terminal. [Figure 11] FIG. 11 is a block configuration diagram for explaining the configuration of second modification of an intercom system including the above-mentioned information terminal. DETAILED DESCRIPTION OF EMBODIMENTS

[0017] The embodiments and modifications described below are merely examples of the present disclosure, and the present disclosure is not limited to the embodiments and modifications. Various changes other than the following embodiments and modifications can be made according to design and the like without departing from the technical idea of the present disclosure.

[0018] (Embodiment) Hereinafter, an intercom system 1 including an information terminal 10 according to the present embodiment will be described with reference to FIGS. 1A to 9. The information terminal 10 operates as, for example, an intercom device.

[0019] (1) Overview The information terminal 10 according to this embodiment will be described below.

[0020] The information terminal 10 according to this embodiment is applied to the intercom system 1, as shown in Figure 2. The intercom system 1 is applied to, for example, an apartment building 5. The intercom system 1 according to this embodiment includes the information terminal 10. In this embodiment, the intercom system 1 includes a plurality of (two in Figure 2) information terminals 10, each of which operates as an intercom device. The intercom system 1 further includes a lobby intercom 20 (intercom entrance device), a control device 30, and a plurality of (two in Figure 2) entrance units 40 (intercom entrance devices; see Figure 2). In other words, the intercom system 1 includes the information terminal 10 and other terminals (lobby intercom 20 and / or entrance units 40).

[0021] In the intercom system 1, each of the multiple information terminals 10 and the lobby intercom 20 communicate via the control device 30. Furthermore, in the intercom system 1, there is a one-to-one correspondence between the multiple information terminals 10 and the multiple door station units 40. Note that the intercom system 1 according to this embodiment may be applied to detached houses in addition to apartment buildings 5. Alternatively, the intercom system 1 may be applied to non-residential facilities such as offices, shops, schools, or nursing homes.

[0022] In other words, the intercom system 1 may be applied to a facility or a collective facility. If the "facility" is a "residential" facility, it may correspond to a detached house or a dwelling unit in a multi-unit dwelling 5, and if it is a "non-residential" facility, it may correspond to an office, shop, school, or nursing home, etc. Furthermore, a "collective facility" may be a multi-unit dwelling 5 consisting of multiple dwelling units (private units), a non-residential collective facility consisting of multiple offices or shops (private units), or a mixed-use facility where, for example, the lower floors are a non-residential collective facility and the upper floors are a multi-unit dwelling.

[0023] Each of the multiple information terminals 10 is, for example, a dwelling unit terminal (intercom master unit) installed in each of the multiple dwelling units E2 included in the apartment building 5. Each information terminal 10 is assumed to be a residential information panel that functions as a room master unit installed in, for example, a room within each dwelling unit E2.

[0024] Each information terminal 10 is connected to the control device 30 via the second main line 62, the branch line 63, and the branch switch 50. Each information terminal 10 is configured to communicate with the lobby intercom 20 via the control device 30 (for example, making calls and transmitting control signals). Furthermore, each information terminal 10 is connected to the corresponding door station 40 via the connection line 64. Each information terminal 10 is configured to communicate with the corresponding door station 40 (for example, making calls and transmitting control signals).

[0025] The lobby intercom 20 is installed, for example, in the common entrance (lobby) E1 of the apartment building 5. The lobby intercom 20 is connected to the control device 30 via the first trunk line 61. The lobby intercom 20 is configured to communicate with each information terminal 10 via the control device 30 (for example, making calls and transmitting video signals). The lobby intercom 20 is mounted, for example, on the wall of the common entrance E1. When the lobby intercom 20 transmits a video signal to the information terminal 10, the information terminal 10 can display video (images).

[0026] In short, the information terminal 10 according to this embodiment includes a communication unit (first communication unit 11, third communication unit 19; see Figure 1A). The communication unit (first communication unit 11, third communication unit 19) performs communication, including voice calls, with other terminals (corresponding in this case to the lobby intercom 20 and the entrance unit 40).

[0027] The control device 30 is installed, for example, in the management room E3 of the apartment building 5. The control device 30 is connected to the lobby intercom 20 via a first trunk line 61 and to each information terminal 10 via a second trunk line 62. In other words, the control device 30 is configured to relay communication between each information terminal 10 and the lobby intercom 20.

[0028] Each of the multiple door station units 40 is installed, for example, at the outer entrance of dwelling unit E2 in apartment building 5. Each door station unit 40 is connected to a corresponding information terminal 10 via a connection line 64. Each door station unit 40 is configured to communicate with the corresponding information terminal 10 (for example, making phone calls, transmitting video signals, etc.).

[0029] In this embodiment, the first main line 61, the second main line 62, the branch line 63, and the connecting line 64 are all twisted pair wires. In other words, although the first main line 61, the second main line 62, the branch line 63, and the connecting line 64 are actually composed of two wires, they are represented as one wire in the diagram. At least one of the first main line 61, the second main line 62, the branch line 63, and the connecting line 64 may be a wire other than a twisted pair wire.

[0030] The information terminal 10 according to this embodiment acquires the voice of a resident in dwelling unit E2 (hereinafter sometimes simply referred to as "user") and performs speech recognition on the acquired voice. That is, as shown in Figure 1A, the information terminal 10 is equipped with a speech recognition unit 182. The speech recognition unit 182 performs speech recognition based on the voice input to the microphone 131. It is assumed that the microphone 131 is provided on the information terminal 10 itself.

[0031] The information terminal 10 controls the operation of the intercom system 1 based on the results of voice recognition. In other words, the information terminal 10 further includes a control processing unit 183, as shown in Figure 1A. The control processing unit 183 performs control based on the voice recognition results of the voice recognition unit 182. Specifically, the control processing unit 183 performs control based on the control commands recognized by the voice recognition unit 182 (for example, control commands that were successfully recognized by voice).

[0032] In short, the information terminal 10 is configured to be voice-operated. For example, if the information terminal 10 determines that the voice of a user in dwelling unit E2 contains a keyword (command word) for a control command to control the opening and closing of the entrance door 200 (door) located at the common entrance E1 (successful voice recognition), it will perform control according to the command word. Specifically, if the information terminal 10 detects "Unlock it!" as a control command through voice recognition by the voice recognition unit 182, it will perform unlocking control of the electric lock 201 to open the entrance door 200 (see Figure 2). Here, the entrance door 200 is configured to be opened and closed by the electric lock 201 (see Figure 2).

[0033] Incidentally, when using voice recognition in intercom devices or systems, improvements to the processing speed and accuracy of voice recognition, as well as improvements to user convenience, are desired.

[0034] Therefore, in the information terminal 10 according to one embodiment of this disclosure, the voice recognition unit 182 is configured to perform voice recognition using either the first method or a second method which is simpler than the first method. The voice recognition unit 182 performs voice recognition using the first method in standby mode and uses the second method when in call mode with another terminal (lobby intercom 20 or door station 40). In other words, the voice recognition unit 182 of the information terminal 10 uses the first method of voice recognition and the second method of voice recognition depending on whether it is in standby mode or call mode. In the above embodiment, when in call mode with another terminal (lobby intercom 20 or door station 40), voice recognition is performed using the second method which is simpler than the first method. Therefore, the processing speed related to voice recognition in call mode is likely to improve, and user convenience may be improved. As a result, the information terminal 10 has the advantage of being able to improve its voice recognition function.

[0035] The term "standby state" refers to a state in which the device is neither in a call with another terminal, nor in a call state where the information terminal 10 is receiving a call from another terminal, nor is it in a state where a predetermined operating mode is being executed. As will be explained in more detail later, a "predetermined operating mode" is, for example, the operating mode of the security set, the operating mode of the door monitor confirmation, and the operating mode of recording playback. In other words, the "standby state" refers to a state in which, for example, only the voice recognition unit 182 (especially the first voice recognition unit 1821 that performs trigger voice recognition) and the microphone 131 are active.

[0036] Furthermore, one embodiment of the present disclosure is a control method for controlling an information terminal 10. The control method includes a speech recognition step, a control processing step, and a communication step. In the speech recognition step, speech recognition processing is performed based on the voice input to the microphone 131. In the control processing step, control is performed based on the speech recognition result of the speech recognition step. In the communication step, communication is performed, including a conversation with other terminals (lobby intercom 20, door station 40). The speech recognition processing is performed using a first method or a second method which is simpler than the first method of speech recognition. In the standby state, speech recognition is performed using the first method, and in the conversation state with other terminals, speech recognition is performed using the second method. The control method according to the above embodiment has the advantage of being able to improve the function of speech recognition. This control method is used on a computer system (information terminal 10). In other words, this control method can also be implemented in a program. A program according to one embodiment of the present disclosure is a program for causing one or more processors to execute the control method according to the above embodiment. The program may be recorded on a computer-readable, non-temporary storage medium.

[0037] Furthermore, in the information terminal 10 according to another embodiment of this disclosure, the types of control commands recognized by the speech recognition unit 182 are set according to the operating state of the device. Therefore, in each operating state of the information terminal 10, when the speech recognition unit 182 identifies which control commands (command words) are included in the voice input to the microphone 131, the possibility of searching by referencing all control commands can be reduced. As a result, the processing speed or accuracy of speech recognition may be improved. Consequently, the information terminal 10 according to the other embodiment has the advantage of being able to improve the function of speech recognition.

[0038] Another form of the control method described herein is a control method for controlling an information terminal 10. The control method includes a speech recognition step, a control processing step, and a communication step. In the speech recognition step, speech recognition processing is performed based on the voice input to the microphone 131. In the control processing step, control is performed based on the control command recognized in the speech recognition step (for example, a control command that was successfully recognized). In the communication step, communication is performed, including a conversation with other terminals (lobby intercom 20, door station 40). The type of control command recognized in the speech recognition step is set according to the operating state of the information terminal 10. The control method described above has the advantage of being able to improve the function of speech recognition. This control method is used on a computer system (information terminal 10). In other words, this control method can also be implemented in a program. A program described in another aspect of the disclosure is a program for causing one or more processors to execute the control method described above. The program may be recorded on a computer-readable non-temporary recording medium.

[0039] In this embodiment, as an example, the information terminal 10 is assumed to be a dwelling unit terminal (intercom master unit), and the "other terminal" referred to in this disclosure is assumed to be a lobby intercom 20 or a door station 40. However, the "other terminal" may also be a manager's room master unit installed in the facility's manager's room (here, for example, management room E3). Furthermore, the information terminal 10 is not limited to a dwelling unit terminal (intercom master unit). For example, the information terminal 10 may be a manager's room master unit, in which case the "other terminal" may be a dwelling unit terminal (intercom master unit) or a lobby intercom 20. Also, the information terminal 10 may be a lobby intercom 20 or a door station 40.

[0040] In the following, unless otherwise specified, the term "other terminals" refers to either the lobby intercom 20 or the entrance unit 40.

[0041] (2) Composition (2-1) Information terminal As shown in Figure 1A, the information terminal 10 includes a first communication unit 11, a second communication unit 12, a sound acquisition unit 13, an operation unit 14, a sound output unit 15, a display unit 16, a storage unit 17, a processing unit 18, and a third communication unit 19.

[0042] The information terminal 10 (for example, the processing unit 18) has, for example, a microcomputer having a processor and memory. The microcomputer functions as the information terminal 10 by executing a program stored in memory through the processor. The program executed by the processor is pre-recorded in the microcomputer's memory, but it may also be provided by being recorded on a non-temporary recording medium such as a memory card, or by being provided via a telecommunications line such as the Internet.

[0043] The first communication unit 11 (communication unit) is a communication interface for communicating with the lobby intercom 20 (its communication unit 21; see Figure 2). That is, the first communication unit 11 performs communication, including voice calls, with other terminals (lobby intercom 20). The first communication unit 11 is connected to the control device 30 via the second trunk line 62, branch line 63, and switch 50. The first communication unit 11 transmits voice signals and control signals, etc., to the lobby intercom 20 via the control device 30. Furthermore, the first communication unit 11 receives call signals, voice signals, and video signals, etc., from the lobby intercom 20 via the control device 30.

[0044] The second communication unit 12 is a communication interface for communicating with the outside (in this case, the server 70). The second communication unit 12 is connected to the server 70 via the network NT1. The second communication unit 12 may receive various information from the server 70. In addition, the control device 30 may be connected to the network NT1 and the server 70 via a gateway or the like, and the information terminal 10 may be connected to the server 70 via the branch line 63 and the control device 30.

[0045] The third communication unit 19 (communication unit) is a communication interface for communicating with the door station 40. That is, the third communication unit 19 performs communication, including voice calls, with other terminals (door station 40). The third communication unit 19 is connected to the door station 40 via a connecting line 64. The third communication unit 19 transmits voice signals and control signals to the door station 40 and receives call signals, voice signals, and video signals from the door station 40.

[0046] The information terminal 10 can take on at least three operating states depending on the communication status with the lobby intercom 20 and the door station 40. The operating states that the information terminal 10 of this embodiment can take on include a call state, a conversation state, and a standby state. The operating states that the information terminal 10 of this embodiment can take on further include a mode operating state. The call state is the state in which the information terminal 10 has received a call signal from the lobby intercom 20 or the door station 40 and is receiving a call. The conversation state is the state in which communication has been established with the lobby intercom 20 or the door station 40 in a conversational manner. The mode operating state is the state in which a predetermined operating mode is being executed. The predetermined operating modes are, as described above, the operating mode for the security set, the operating mode for checking the door monitor, and the operating mode for recording and playback. The standby state is a state that is neither the conversation state, the call state, nor a mode operating state.

[0047] The operation of equipment control such as air conditioners, and the operation of scene control, as described later, are instantaneous operations for the information terminal 10, involving the transmission of instruction signals to external devices to instruct control. These operations are unlikely to be maintained for a relatively long period of time (e.g., more than one minute). However, even the operation of equipment control such as air conditioners, and the operation of scene control, may correspond to the above-described mode operation state if the information terminal 10 is in the state of performing the operation for a relatively long period of time. For example, if scene control includes the execution of a recording and playback operation mode (see "Operation Example 6" described later), the operation mode of scene control (recording and playback) may also be one of the predetermined operation modes described above.

[0048] The sound acquisition unit 13 acquires sound, including at least the user's voice, and outputs sound information (voice data) related to the acquired sound to the processing unit 18. The sound acquisition unit 13 has a microphone 131 (see Figure 1A). The sound acquisition unit 13 converts ambient sound, including the voice of the user located in front of the information terminal 10, which is input to the microphone 131, into an analog sound signal (sound information) and outputs it to the processing unit 18.

[0049] The sound output unit 15 has a speaker 151. When the information terminal 10 is in a state where it can communicate with the lobby intercom 20 (communication state), the speaker 151 outputs sound (including the voice of a visitor, etc.) based on the voice signal (sound data) transmitted from the lobby intercom 20. When the information terminal 10 is in a state where it can communicate with the door station 40 (communication state), the speaker 151 outputs sound (including the voice of a visitor, etc.) based on the voice signal (sound data) transmitted from the door station 40. In addition, depending on the status of the information terminal 10, the speaker 151 outputs various voice messages and electronic sounds stored in the storage unit 17, for example.

[0050] In Figure 1A, two microphones 131 are shown. That is, the sound acquisition unit 13 has two microphones 131 (first microphone 1311 and second microphone 1312).

[0051] The first microphone 1311 and the second microphone 1312 are positioned, for example, on the upper left and right sides when viewed from the front of the information terminal 10. In particular, the first microphone 1311 is located further away from the speaker 151 than the second microphone 1312. In other words, the first microphone 1311 is a far-end microphone.

[0052] In this embodiment, the information terminal 10 utilizes, for example, the user's voice input to the first microphone 1311 as voice data for calls when making a call with another terminal. The voice input to the first microphone 1311 is mainly processed for call audio and then transmitted to the other terminal. In other words, by using the voice from the far-end first microphone 1311, rather than the second microphone 1312, as voice data for calls, it suppresses the leakage of the (visitor's) voice output from the speaker 151 to the microphone 131.

[0053] On the other hand, the information terminal 10 of this embodiment utilizes, for example, the user's voice input to the first microphone 1311 and the second microphone 1312 as voice data for speech recognition. In particular, the processing unit 18 has a function to perform beamforming processing to enhance the directivity of the sound reception direction using the two microphones 131. The processing unit 18 estimates from the user's voice input from the first microphone 1311 and the second microphone 1312 whether the user is speaking to the information terminal 10 from the left or the right. Then, based on the estimation result, the processing unit 18 performs processing to increase the sensitivity to voice signals from a specific direction.

[0054] The operation unit 14 is configured to accept user input. As shown in Figure 1B, the operation unit 14 includes an answer operation unit 141 (call operation unit), an end operation unit 142, and an unlock operation unit 143. The operation unit 14 also includes buttons (home operation unit) for displaying the home screen on the display unit 16.

[0055] In other words, the information terminal 10 is equipped with an answering operation unit 141 that receives an answering operation for answering a call to the information terminal 10 from another terminal. The answering operation unit 141 is an operation unit (button) for initiating communication (a conversation with a visitor, etc.) with the lobby intercom 20 or the entrance unit 40 in response to a call from the lobby intercom 20 or the entrance unit 40. That is, for example, if the answering operation unit 141 is operated while the first communication unit 11 is in a state where it has received a call signal to call a resident (call state), voice communication becomes possible between the lobby intercom 20 and the information terminal 10.

[0056] In other words, the information terminal 10 further includes at least one (both in this embodiment) of an termination operation unit 142 and an unlocking operation unit 143. The termination operation unit 142 receives termination operations to end a call with another terminal. The unlocking operation unit 143 receives unlocking operations to unlock an electric lock 201 installed on a door used for entering and exiting a facility (here, an apartment building 5) (here, an entrance door 200 provided at the common entrance E1; see Figure 2). That is, for example, when the first communication unit 11 is in a call state with the lobby intercom 20, if the termination operation unit 142 is operated, the voice call with the lobby intercom 20 is terminated. Also, for example, when the first communication unit 11 is in a call state with the lobby intercom 20, if the unlocking operation unit 143 is operated, an unlocking signal is transmitted from the information terminal 10 to the electric lock 201 via the control device 30 to cause it to unlock. As a result, the electric lock 201 is unlocked, and the visitor can open the front door 200 and enter beyond the common entrance E1. In this embodiment, it is assumed that even if the unlocking operation unit 143 is operated and the electric lock 201 is unlocked, the information terminal 10 does not automatically terminate the voice call with the lobby intercom 20, however, the voice call may be automatically terminated in response to the unlocking of the electric lock 201. Alternatively, the lobby intercom 20 may have a separate communication unit (electric lock communication unit) from the communication unit 21, and the electric lock communication unit may be connected to the electric lock 201. In this case, when the information terminal 10 performs unlocking control, an unlocking signal is output from the lobby intercom 20 to the electric lock 201, and the electric lock 201 is unlocked.

[0057] In this embodiment, as an example, the response operation unit 141, the termination operation unit 142, the unlock operation unit 143, and the home operation unit are implemented by capacitive touch sensors arranged around the display unit 16 on the front of the housing of the information terminal 10. That is, the user can input an operation simply by lightly touching the response operation unit 141, the termination operation unit 142, the unlock operation unit 143, and the home operation unit with their fingertips. However, the response operation unit 141, the termination operation unit 142, the unlock operation unit 143, and the home operation unit are not limited to being implemented by touch sensors, and at least some of them may be implemented by push-button type switches.

[0058] The display unit 16 is, for example, a liquid crystal display. The display unit 16 is configured to display video. When the information terminal 10 is in a state where it can communicate with the lobby intercom 20, the display unit 16 displays video captured by the lobby intercom 20, which is the communication target. When the information terminal 10 is in a state where it can communicate with the door station 40, the display unit 16 displays video captured by the door station 40, which is the communication target. If the information terminal 10 is equipped with a touch panel display, the touch panel display may combine the functions of the display unit 16 and, for example, some functions of the operation unit 14. The display unit 16 also displays a home screen, a settings screen, and screens corresponding to each operating state.

[0059] The memory unit 17 includes a first memory unit 171 and a second memory unit 172. The first memory unit 171 and the second memory unit 172 are each composed of read-write memory. The first memory unit 171 and the second memory unit 172 are, for example, flash memory. The memory unit 17 stores, for example, dictionary files used for speech recognition (trigger dictionary D0, command dictionary D1; see Figure 1A). The dictionary files may be downloaded from the server 70 and stored. The dictionary files include information relating to at least one of the following: the type of language that the speech recognition unit 182 can recognize (Japanese, English, etc.), the expression method in the same language (dialect, accent, etc.), and the speaker in the same language (male, female, child, elderly, etc.). Hereinafter, a dictionary file based on at least one of the type of language that the speech recognition unit 182 can recognize, the expression method in the same language, and the speaker in the same language will be referred to as a dictionary file based on language information. In addition to the dictionary file, the memory unit 17 also stores information about the screen to be displayed on the display unit 16, information about voice messages, and various setting information.

[0060] Trigger dictionary D0 is a dictionary containing trigger words. A trigger word is a predetermined keyword that triggers the detection of command words of a control command. Command dictionary D1 is a dictionary containing command words of a control command. Command dictionary D1 includes a first dictionary D11 and a second dictionary D12, as shown in Figure 1A.

[0061] In this embodiment, the trigger word is set as a keyword predetermined by the information terminal 10 or the provider of the dictionary file (for example, the manufacturer of the information terminal 10). The trigger word is, for example, the product name of the information terminal 10 or its abbreviation, and it is desirable that it has a relatively small number of characters. The trigger word may also be set by the user of the information terminal 10 to their preferred word as appropriate on the settings screen of the information terminal 10. In Figure 4, for example, in the resident's statement "○○, respond!", "○○" corresponds to the trigger word, and "Respond!" following "○○" corresponds to the command word of the control command.

[0062] In this embodiment, as an example, the trigger dictionary D0 is stored in the first storage unit 171, and the command dictionary D1 (first dictionary D11 and second dictionary D12) is stored in the second storage unit 172.

[0063] As shown in Figure 1A, the processing unit 18 includes a voice processing unit 181, a voice recognition unit 182, a control processing unit 183, a display processing unit 184, and a call processing unit 185.

[0064] The audio processing unit 181 acquires the analog sound signal input from the first microphone 1311 and output by the sound acquisition unit 13 for voice communication and converts it into a digital sound signal. The audio processing unit 181 is also configured to perform filtering and other processing on the sound signal input from the first microphone 1311 and acquired by the sound acquisition unit 13. The audio processing unit 181 includes, for example, an echo canceller. The echo canceller suppresses or removes echoes from the sound signal output from the microphone 131. The audio processing unit 181 is also configured to acquire the sound signals input from the first microphone 1311 and the second microphone 1312 and output by the sound acquisition unit 13 for voice identification and perform the beamforming processing described above.

[0065] The speech recognition unit 182 performs speech recognition based on the sound input to the microphone 131 (in this embodiment, the first microphone 1311 and the second microphone 1312). In other words, the speech recognition unit 182 performs speech recognition processing based on the sound acquired by the sound acquisition unit 13. The speech recognition unit 182 performs speech recognition processing on the sound processed by the speech processing unit 181 using dictionary files (trigger dictionary D0, command dictionary D1) stored in the storage unit 17.

[0066] The voice recognition unit 182 of this embodiment performs voice recognition using either the first method or a second method which is simpler than the first method. The voice recognition unit 182 performs voice recognition using the first method in standby mode and uses the second method when in a call with another terminal.

[0067] Here, the "first method" is a method in which trigger speech recognition is performed to determine whether or not a trigger word is included in the audio input to the microphone 131, and if a trigger word is included, command speech recognition is performed. Command speech recognition is speech recognition of control commands related to the control performed by the control processing unit 183, that is, it recognizes whether or not a particular command word is included.

[0068] On the other hand, the "second method" is a method that performs command speech recognition on the sound input to the microphone 131 without performing trigger speech recognition. In short, the second method is simpler than the first method because it omits the process of trigger speech recognition.

[0069] Furthermore, in this embodiment, the "first method" is a method in which the control processing unit 183 performs voice recognition (first command voice recognition) of control commands (first control commands) related to predetermined controls performed in the standby state, based on the first dictionary D11. The "predetermined controls" referred to here are mainly controls "other than" those related to calls, which will be described later.

[0070] On the other hand, the "second method" is a method in which the control processing unit 183 performs speech recognition (second command speech recognition) of control commands (second control commands) related to specific controls performed during a call, based on the second dictionary D12. The "specific controls" referred to here are mainly the control related to calls, which will be described later.

[0071] In other words, in this embodiment, the dictionary file used for speech recognition is changed between the standby state and the call state, using the first dictionary D11 and the second dictionary D12.

[0072] Furthermore, the number of first control commands registered in the first dictionary D11 is set to be greater than the number of second control commands registered in the second dictionary D12. In short, the second method is simpler than the first method because it refers to (searches) within the second dictionary D12, which has fewer registered command words than the first dictionary D11.

[0073] In other words, the voice recognition unit 182 is configured to perform voice recognition processing not only when communication is in progress (call state) between the information terminal 10 and the lobby intercom 20 (or door station 40), but also when the information terminal 10 is in standby mode.

[0074] Furthermore, the voice recognition unit 182 is configured to perform voice recognition processing even when a call is received from another terminal, and in the mode operation state described above.

[0075] In short, the information terminal 10 is capable of receiving sound from the microphone 131 even when it is not in a call with another terminal.

[0076] Here, the speech recognition unit 182 includes a first speech recognition unit 1821 that performs trigger speech recognition and a second speech recognition unit 1822 that performs command speech recognition (see Figure 1A). In other words, in this embodiment, the part (circuit) that performs speech recognition processing related to trigger words and the part (circuit) that performs speech recognition processing related to command words are implemented separately. In the standby state of the information terminal 10, the first speech recognition unit 1821 and the sound acquisition unit 13, which mainly perform trigger speech recognition, are in an active state (started state, i.e., a state that consumes power), and monitor whether speech including trigger words is input to the microphone 131. Conversely, in the standby state of the information terminal 10, the second speech recognition unit 1822 that performs command speech recognition and the control processing unit 183 (hereinafter also referred to as "second speech recognition unit 1822, etc.") are in an inactive state (non-started state, i.e., a state that does not consume power, or a sleep state with reduced power consumption).

[0077] The first speech recognition unit 1821 monitors the sound processed by the speech processing unit 181 by referencing (searching) the trigger dictionary D0 through trigger speech recognition and determines whether or not a trigger word is included. When the first speech recognition unit 1821 determines that a trigger word is included (trigger word detection), the information terminal 10 activates the second speech recognition unit 1822, etc., and makes the control command available for speech recognition.

[0078] For example, a response command (for example, the command word "respond") is set as a control command related to the control performed by the control processing unit 183. The response command is a command for answering a call to the information terminal 10 from another terminal. In this embodiment, the user needs to utter a trigger word immediately before the response command. When the information terminal 10 determines that the trigger word is included, it activates the second speech recognition unit 1822, etc., and makes the response command speech-recognizable.

[0079] In addition to detecting trigger words, the information terminal 10 may also activate the second voice recognition unit 1822, etc., by receiving a call from the lobby intercom 20 or the entrance unit 40, or by receiving a touch operation on the operation unit 14 or the display unit 16.

[0080] The second speech recognition unit 1822 monitors the sound processed by the speech processing unit 181 by performing command speech recognition that refers to (searches) the first dictionary D11 or the second dictionary D12, and determines whether or not a command word is included. If the second speech recognition unit 1822 determines that a certain command word is included (command word detection), it outputs that command word (control command) to the control processing unit 183.

[0081] The second speech recognition unit 1822 remains in an inactive state until the first speech recognition unit 1821 recognizes that a trigger word is included in the speech. In other words, in this embodiment, the parts (processing circuits) that perform trigger speech recognition and command speech recognition are separated, and the second speech recognition unit 1822 is kept in an inactive state until it is recognized that a trigger word is included. Then, when the second speech recognition unit 1822 recognizes that a trigger word is included in the speech recognition unit 1822 in an active state that enables command speech recognition. As a result, the processing speed of the speech recognition unit 1822 can be improved. In addition, the power consumption of the second speech recognition unit 1822 can be suppressed.

[0082] The "inactive state" of the second speech recognition unit 1822 as used here refers to a state in which the second speech recognition unit 1822 is inactive and not consuming power, or is in a sleep state with reduced power consumption. However, the "inactive state" of the second speech recognition unit 1822 may also refer to a state in which the second speech recognition unit 1822 is active but has suspended command speech recognition processing.

[0083] There are exceptions to the rule that the second speech recognition unit 1822 is disabled in the standby state until it is recognized that a trigger word is included. For example, if a control request is received from the user through a touch operation on the direct operation unit 14, the second speech recognition unit 1822 may become capable of command speech recognition (enabled state) even without a trigger word.

[0084] The control processing unit 183 performs control based on the speech recognition results of the speech recognition unit 182. In other words, the control processing unit 183 performs control based on the control commands recognized by the speech recognition unit 182 (for example, control commands that were successfully recognized). Here, the control processing unit 183 executes the control associated with the speech recognition results (control commands) obtained from the second speech recognition unit 1822. The control processing unit 183 also notifies whether speech recognition was successful or unsuccessful by sound or display (for example, output of an electronic sound or voice message, display by lighting or flashing an indicator light, or display of a message on the screen of the display unit 16).

[0085] In short, when the information terminal 10 is in standby mode, it refers to the trigger dictionary D0 to detect a trigger word, switches the second speech recognition unit 1822 etc. from inactive to active, and refers to the first dictionary D11 to perform speech recognition of the command word (first method). When the information terminal 10 is in call mode, it refers to the second dictionary D12 to perform only speech recognition of the command word (second method). When the information terminal 10 receives a call, it switches from the first dictionary D11 to the second dictionary D12. When the information terminal 10 is in call mode, it detects a trigger word, switches the second speech recognition unit 1822 etc. from inactive to active, and refers to the second dictionary D12 to perform speech recognition of the command word. After that, the information terminal 10 switches from the first method to the second method to perform speech recognition. In other words, the timing of switching the reference from the first dictionary D11 to the second dictionary D12, the timing of switching the second speech recognition unit 1822 etc. from inactive to active, and the timing of switching from the first method to the second method are different. However, at least two of these three timings may be the same.

[0086] When a response command (control command) to a call from another terminal is recognized by voice (successful voice recognition), the control processing unit 183 controls the operation of the information terminal 10 to initiate communication (call) with the other terminal. For example, the control processing unit 183 controls the first communication unit 11 (or third communication unit 19) to establish communication so that a call can be made between the information terminal 10 and the lobby intercom 20 (or entrance unit 40).

[0087] The voice recognition unit 182 performs only command voice recognition in the second method until communication (call) between the information terminal 10 and other terminals ends, or until the electric lock 201 of the entrance door 200 (see Figure 2) is unlocked. In other words, the voice recognition unit 182 does not perform trigger voice recognition until communication (call) ends, or until the electric lock 201 is unlocked.

[0088] Furthermore, if the voice recognition unit 182 is in the second voice recognition mode and no voice input is received by the microphone 131 for a predetermined period of time (e.g., several minutes), it will automatically switch to the first voice recognition mode. At the same time, the information terminal 10 will also automatically terminate any calls with other terminals. As a result, the system can switch to the first mode even without user voice input, reducing the possibility of the system remaining in the second mode, improving user convenience, and enabling improvements to the voice recognition function.

[0089] The call processing unit 185 outputs the call audio data (sound information) acquired by the sound acquisition unit 13 to a device operated by the person on the other end of the call. The information terminal 10, for example, uses the user's voice input to the first microphone 1311 as call audio data when making a call with another terminal. The voice input to the first microphone 1311 is mainly processed for call audio in the call processing unit 185 and then transmitted to the other terminal making the call. In other words, by using the voice from the far-end first microphone 1311 instead of the second microphone 1312 as call audio data, it is suppressed that the voice (of the visitor) output from the speaker 151 leaks into the microphone 131. Specifically, the call processing unit 185 transmits the sound signal, which has had noise suppressed or removed by the voice processing unit 181, to the lobby intercom 20 (or door station 40) via the first communication unit 11 (or third communication unit 19). For example, when the information terminal 10 is communicating with the lobby intercom 20, the call processing unit 185 transmits a signal of sound data, which has had noise suppressed or removed by the voice processing unit 181, to the lobby intercom 20 via the first communication unit 11.

[0090] The display processing unit 184 performs processing to display images captured by the intercom entrance device (lobby intercom 20, entrance unit 40), which is the communication target, on the display unit 16. The display processing unit 184 also performs processing to display the home screen, settings screen, and screens corresponding to each operating status.

[0091] Table 1 below shows a list of specific control commands (command words) and the control events (operation events) associated with each control command. The control commands are associated with operation events and stored in the memory unit 17. The "Dictionary" column on the far right of Table 1 shows the dictionary in which the corresponding command words are registered.

[0092] [Table 1]

[0093] The command words shown in Table 1 are merely examples and are not limited to them. From Table 1, it is easy to see that the number of command words in the second dictionary D12 is less than the number of command words in the first dictionary D11. Also from Table 1, it is easy to see that the operation events associated with the command words in the second dictionary D12 are limited to intercom control-related events, and in particular to operation events that are highly related to "calls". Conversely, it is easy to see that the operation events associated with the command words in the first dictionary D11 are limited to operation events for equipment control other than the intercom device, scene control, and function control within intercom control that are less related to "calls". In the following, operation events that are highly related to "calls" may also be referred to as "call-related operation events". Call-related operation events include operation events related to receiving calls from other terminals, answering calls, ending calls, and unlocking the electric lock 201 of the entrance door 200 (see Figure 2).

[0094] In the information terminal 10, the operation control performed by voice input (voice operation) of the control commands in Table 1 can also be performed by touch operation input to the operation unit 14 or the display unit 16.

[0095] When the control processing unit 183 receives voice input containing control commands as shown in Table 1 and the second voice recognition unit 1822 recognizes the voice (successful voice recognition), it identifies the corresponding operation event and controls the target device (including the user). If the operation event is related to intercom control, the target device for control will be the user (information terminal 10) or the electric lock 201 of the entrance door 200 (see Figure 2).

[0096] For example, while information terminal 10 is communicating (on a call) with lobby intercom 20, information terminal 10 receives a voice message from the user containing the command word "end" as a control command (voice input). In this case, the second voice recognition unit 1822 determines that the voice processed by the voice processing unit 181 contains the command word "end". The control processing unit 183 controls the operation to disconnect the call between itself and lobby intercom 20.

[0097] Alternatively, instead of voice input, the control processing unit 183 controls the operation to disconnect the call when the termination operation unit 142 receives a touch operation from the user.

[0098] Furthermore, for example, while the information terminal 10 is communicating with the lobby intercom 20 (on a call), the information terminal 10 receives voice from the user as a control command, which includes the command word "unlock" (voice input). In this case, the second voice recognition unit 1822 determines that the voice processed by the voice processing unit 181 includes the command word "unlock". The control processing unit 183 controls the unlocking operation of the electric lock 201 of the entrance door 200 for entering the room from the common entrance E1. For example, the control processing unit 183 controls the control device 30 so that the electric lock of the entrance door 200 for entering the room from the common entrance E1 performs the unlocking operation.

[0099] Alternatively, instead of voice input, the control processing unit 183 also controls the unlocking of the electric lock 201 of the entrance door 200 when the unlocking operation unit 143 receives a touch operation from the user.

[0100] In addition to its intercom function, the information terminal 10 also has a disaster prevention function to activate a fire alarm when a fire is detected, a security function to activate an alarm when an intruder is detected, and a function to display various lifestyle information. Furthermore, the information terminal 10 has a communication function to communicate with an external source (in this case, the server 70). For example, the "security set" shown in Table 1 refers to a setting (set) that turns on the alert mode using the security function described above.

[0101] Regarding equipment control and scene control, for example, it is assumed that the information terminal 10 is connected to the HEMS (Home Energy Management System) controller installed in each dwelling unit E2 in a communicative manner. The HEMS controller is connected in a way that allows it to control electrical equipment (air conditioners, lighting fixtures, air purifiers, range hoods, etc.) as shown in Table 1. In short, regarding equipment control and scene control, the equipment to be controlled can be any electrical equipment that the HEMS controller can control. When the information terminal 10 receives control commands related to equipment control or scene control via voice input, it can control the corresponding electrical equipment via the HEMS controller.

[0102] In other words, the information terminal 10 is configured to allow users to easily switch between two patterns: a first pattern in which a trigger word is spoken, followed immediately by a command word (for example, within a fraction of a second to a few seconds), and a second pattern in which a command word is spoken alone. The information terminal 10 then executes control for non-call-related operation events within device control, scene control, and intercom control (such as security system activation or recording playback) when it receives voice commands using the first pattern. On the other hand, for call-related operation events within intercom control, the information terminal 10 executes control when it receives voice commands using either the first or second pattern. In other words, for call-related operation events, even if a trigger word is included, the information terminal 10 simply does not perform trigger voice recognition; it will execute control if the correct command word is included afterward. However, if a user performs a voice command using the first pattern while on a call, there is a high possibility that the trigger word will be revealed to the other party on the call (for example, a visitor), which could reduce security. Furthermore, for the person on the other end of the call, hearing a trigger word abruptly during a call can seem unnatural and may cause discomfort. Therefore, as a way of handling the information terminal 10, users are advised to use the second pattern of voice operation, in which they utter the command word alone while on a call. The voice recognition unit 182 may also perform trigger voice recognition processing even while on a call, and if it detects a trigger word, the information terminal 10 may display a message on the display unit 16 indicating that users should refrain from uttering the trigger word while on a call.

[0103] Furthermore, in this embodiment, the control commands recognized by the speech recognition unit 182 include command words that are set to be the same word across multiple operating states. For example, in the example in Table 1, the command word "end" is set to be the same word in the call state (and calling state) and the mode operating state in which a predetermined operation is being performed. This command word "end" is registered in both the first dictionary D11 and the second dictionary D12 (see Table 1). The control processing unit 183 performs different controls depending on the operating state of its own device, even if the control command recognized by the speech recognition unit 182 is the same word. Specifically, for example, the control content of the command word "end" differs depending on the operating state of the information terminal 10 when the user speaks it. If the information terminal 10 successfully recognizes the command word "end" while in a call state, it performs control to disconnect the call. However, if it successfully recognizes the command word "end" while in the state of performing a door monitor check, it performs control to terminate the door monitor check operation. Furthermore, if the information terminal 10 successfully recognizes the command word "end" while recording playback is in progress, it will perform control to terminate the recording playback operation. In this way, by setting the same word across multiple operating states, the number of control command words that the user needs to remember can be reduced, improving user convenience.

[0104] The information terminal 10 is configured to allow modification of the dictionary files referenced by voice recognition (trigger dictionary D0, command dictionary D1) and corresponding control events (operation events) by opening a settings screen via touch operation on the operation unit 14 or the display unit 16. Furthermore, it is preferable that the information terminal 10 allows for at least one of the following: addition of the types of languages ​​that can be recognized by voice (Japanese, English, etc.), addition of expression methods within the same language (dialects, etc.), and addition of the speaker type within the same language (male, female, child, elderly, etc.). The information terminal 10 may also update the dictionary files in the storage unit 17 by downloading update data related to the dictionary files from the server 70, for example.

[0105] (2-2) Lobby intercom As shown in Figure 2, the lobby intercom 20 consists of a communication unit 21, a control unit 22, and a call unit. It comprises a 23, a display unit 24, an operation unit 25, a storage unit 26, and an imaging unit 27.

[0106] The lobby intercom 20 has, for example, a microcomputer having a processor and memory. The microcomputer functions as a control unit 22 by executing a program stored in memory through the processor. The program executed by the processor is pre-recorded in the microcomputer's memory, but it may also be provided by being recorded on a non-temporary recording medium such as a memory card, or by being provided via a telecommunications line such as the Internet.

[0107] The communication unit 21 is a communication interface for communicating with the information terminal 10 (specifically, its first communication unit 11). The communication unit 21 is connected to the control device 30 (specifically, its communication unit 31) via the first trunk line 61. The communication unit 21 transmits call signals, voice signals, and video signals, etc., to the information terminal 10 via the control device 30. Furthermore, the communication unit 21 receives voice signals and control signals, etc., from the information terminal 10 via the control device 30. Here, the communication signals from the lobby intercom 20 contain information for identifying the information terminal 10 (for example, address information, etc.). Therefore, only information terminals 10 that have been assigned address information that matches the address information contained in this communication signal can receive the communication signal.

[0108] The control unit 22 is configured to control the communication unit 21, the call unit 23, the imaging unit 27, and the like.

[0109] The communication unit 23 includes a speaker and a microphone and is configured to enable communication with the information terminal 10.

[0110] The display unit 24 is, for example, a liquid crystal display. The display unit 24 is configured to display images captured by the imaging unit 27. The display unit 24 is also configured to display messages to visitors, etc. The messages are, for example, messages to prompt visitors, etc. to speak. The display unit 24 displays messages such as, for example, "Please speak." In this case, a similar voice message may be output (announced) from the speaker of the communication unit 23, or from a speaker provided separately from the speaker of the communication unit 23. Alternatively, the display unit 24 and the speaker may be used in combination. If the lobby intercom 20 is equipped with a touch panel display, the touch panel display may serve as both the display unit 24 and the operation unit 25.

[0111] The control unit 25 is configured to accept operations from, for example, visitors or residents of the apartment building 5. The control unit 25 is an input interface having, for example, multiple push-button switches and a touch panel.

[0112] The storage unit 26 is composed of read-write memory. The storage unit 26 is, for example, flash memory. The storage unit 26 stores, for example, video data of images captured by the imaging unit 27.

[0113] The imaging unit 27 is a camera having an image sensor for capturing images of a subject (such as a visitor). In this embodiment, the imaging area (field of view) of the imaging unit 27 is set in front of the lobby intercom 20. In this embodiment, the imaging unit 27 is a camera that captures video. Furthermore, in this embodiment, the imaging unit 27 is a camera that captures color images. The imaging unit 27 may also be a camera that captures still images (still camera) or a camera that captures monochrome images.

[0114] The image sensor is a two-dimensional image sensor such as a CCD (Charge Coupled Devices) image sensor or a CMOS (Complementary Metal-Oxide Semiconductor) image sensor. The imaging unit 27 forms an image of light from the subject onto the imaging surface (light-receiving surface) of the image sensor using an optical system such as a lens, and converts the light from the subject into an electrical signal at the image sensor. The imaging unit 27 then outputs the output signal from the image sensor as a video signal to the control unit 22.

[0115] (2-3) Control device As shown in Figure 2, the control device 30 includes a communication unit 31, a control unit 32, and a storage unit 33.

[0116] The control device 30 includes, for example, a microcomputer having a processor and memory. The microcomputer functions as a control unit 32 when the processor executes a program stored in memory. The program executed by the processor is pre-recorded in the microcomputer's memory, but it may also be provided on a non-temporary recording medium such as a memory card, or provided via a telecommunication line such as the Internet.

[0117] The communication unit 31 includes a communication interface for communicating with each information terminal 10 and the lobby intercom 20. The communication unit 31 is connected to the lobby intercom 20 via a first trunk line 61 and to each information terminal 10 via a second trunk line 62. In other words, the communication unit 31 is configured to relay communication between each information terminal 10 and the lobby intercom 20. Furthermore, the communication unit 31 is connected to the electric lock 201 of the entrance door 200 and is configured to communicate with the electric lock 201.

[0118] The control unit 32 is configured to control the communication unit 31. The control unit 32 controls the unlocking operation of the electric lock 201 of the entrance door 200 based on instructions from the information terminal 10. For example, the control unit 32 outputs an unlocking signal to the electric lock 201 via the communication unit 31 to instruct it to unlock.

[0119] The storage unit 33 is composed of read-write memory. The storage unit 33 is, for example, flash memory. The storage unit 33 stores, for example, a correspondence table that shows the correspondence between the room number assigned to each dwelling unit E2 and the address information assigned to each information terminal 10. In other words, in the control device 30, the control unit 32 refers to the correspondence table and creates a signal in which the room number included in the signal from the lobby intercom 20 is replaced with the address information of the corresponding information terminal 10, and has the communication unit 31 send this signal to each information terminal 10. Then, in each information terminal 10, the processing unit 18 acquires the information included in the signal if the address information included in the signal received by the first communication unit 11 matches the address information stored in the storage unit 17. Also, in each information terminal 10, the processing unit 18 discards the information included in the signal if the address information included in the signal received by the first communication unit 11 does not match the address information stored in the storage unit 17.

[0120] In this embodiment, the communication unit 31 is connected to the electric lock 201, but the configuration is not limited to this. The control device 30 may have a separate communication unit from the communication unit 31, and this separate communication unit may be connected to the electric lock 201.

[0121] (2-4) Door station Each door station 40 is connected to the corresponding information terminal 10 via a connection line 64, as shown in Figure 2. The door station 40 transmits call signals, voice signals, and video signals to the information terminal 10. Furthermore, the door station 40 receives voice signals and control signals from the information terminal 10.

[0122] (3) Operation The operation of the information terminal 10 will be explained by giving six examples of operation, referring to Figures 3 to 9.

[0123] (3-1) Example of operation 1 The following describes an example of the operation of the information terminal 10 (Operation Example 1) regarding "call-related" operation events within the intercom control operation events, with reference to Figures 3 and 4.

[0124] Figure 3 is a flowchart relating to Operation Example 1. Figure 4 is a conceptual diagram showing a series of operations of the information terminal 10, from standby state, receiving a call (incoming call), answering the call (answering), during the call, unlocking the electric lock 201 (unlocking), ending the call (end call), and returning to standby state, arranged chronologically from left to right along with the screen of the display unit 16. The upper part of Figure 4 shows the operation of the information terminal 10 when the means of operation (means) for the information terminal 10 when the user requests control is a touch operation on the operation unit 14 (or display unit 16) (labeled "button" in Figure 4). On the other hand, the lower part of Figure 4 shows the operation of the information terminal 10 when the means of operation (means) is voice operation to the microphone 131 (labeled "voice" in Figure 4). Figure 3 is a flowchart for the case of voice operation.

[0125] First, "Standby or Home" in Figure 4 indicates that the information terminal 10 is in standby mode. In standby mode, the information terminal 10 displays either a standby screen with the display unit 16 dimmed to conserve power, or the home screen (labeled "Home" in Figure 4) on the display unit 16.

[0126] In standby mode, if, for example, an incoming call is received from the lobby intercom 20 (incoming call), the information terminal 10 will sound a ringtone from the speaker 151 to notify the user of the incoming call. Furthermore, the information terminal 10 will display an incoming call screen on the display unit 16 to indicate that an incoming call is in progress (call status). At the moment an incoming call occurs, the information terminal 10 will switch the dictionary used for voice recognition from the first dictionary D11 to the second dictionary D12, which contains command words corresponding to call-related operation events.

[0127] Here, the user can respond to a call by touching the response operation unit 141 (labeled "call button" in Figure 4) while the information terminal 10 is in the call state. In other words, when the information terminal 10 is in the call state (a state in which it has received a call signal from another terminal), if it receives a touch operation on the response operation unit 141, it switches from the call state to the call state and enables a conversation with the person operating the lobby intercom 20 (answer). The information terminal 10 changes the incoming call screen displayed on the display unit 16 to the call screen (screen transition).

[0128] On the other hand, the user can also respond to a call by using voice commands on the microphone 131 while the information terminal 10 is in call mode. As shown in Figure 4, the user speaks a trigger word, "○○, answer!", and a command word (response command) while in call mode. When the first voice recognition unit 1821 of the information terminal 10 successfully recognizes the trigger word "○○", it activates the second voice recognition unit 1822 and other components. Furthermore, when the second voice recognition unit 1822 successfully recognizes the subsequent spoken "Answer!", the information terminal 10 switches from call mode to call mode, enabling communication with the person operating the lobby intercom 20 (response). At this time, the second voice recognition unit 1822 performs voice recognition by referring to the second dictionary D12, resulting in a fast processing speed for voice recognition. To inform the user that the response command has been received (i.e., that the voice input was successful), the information terminal 10 outputs a "beep" sound from the speaker 151. Furthermore, if the information terminal 10 fails to detect a valid command word within a specified time (e.g., 3 seconds) after detecting a trigger word, it outputs a voice message saying, "Could not recognize the command." In this case, for example, the information terminal 10 returns the second voice recognition unit 1822, etc., to an inactive state.

[0129] Here, as an example, the information terminal 10 switches the speech recognition method from the first method to the second method at the moment it switches from the call state to the call state. In other words, the speech recognition unit 182 switches to the second method of speech recognition by accepting a response operation to the response operation unit 141 (while in the state of speech recognition using the first method). Also, the speech recognition unit 182 switches to the second method of speech recognition by recognizing a response command (successful speech recognition of the response command) while in the state of speech recognition using the first method. In this way, the switch to the second method is achieved by a response operation to the response operation unit 141 or by uttering a response command, thus improving user convenience.

[0130] Refer to the flowchart in Figure 3. Information terminal 10 determines whether it is in a call (call state) or not (step ST1). If it is in a call (step ST1: Yes), it waits for a command word (step ST2). On the other hand, if information terminal 10 is not in a call, such as in standby or calling state (step ST1: No), it waits for a trigger word (step ST3). If information terminal 10 succeeds in voice recognition of the trigger word, that is, if it succeeds in authentication of the trigger word (authentication OK?; step ST4: Yes), it waits for a command word (step ST2). If information terminal 10 is not in a call, unless it succeeds in voice recognition of the trigger word, that is, if it fails to authenticate the trigger word (authentication OK?; step ST4: No), it returns to the original step ST1. Note that this flowchart shows the processing flow considering only voice operation for the sake of clarity, and this does not apply when touch operation is added.

[0131] During a call, the call screen is displayed on the display unit 16. While on a call, the user can unlock the electric lock 201 by touching the unlocking operation unit 143 (labeled "unlock button" in Figure 4) in order to invite a visitor through the front door 200. In other words, when the information terminal 10 receives a touch operation to the unlocking operation unit 143 while on a call, it executes unlocking control for the electric lock 201 via the control device 30 (unlocks it). The information terminal 10 displays a message on the call screen displayed on the display unit 16 indicating that the unlocking control for the electric lock 201 has been executed.

[0132] Alternatively, the user can also unlock the electric lock 201 by using voice commands on the microphone 131 while the information terminal 10 is in a call state. As shown in Figure 4, the user speaks the command word "Unlock it!" (unlock command) while in a call state. The unlock command is a command to unlock the electric lock 201 installed on the door (front door 200) used for entering and exiting the facility (apartment building 5) where the information terminal 10 is installed. When the second voice recognition unit 1822 of the information terminal 10 successfully recognizes the spoken "Unlock it!", it executes unlock control of the electric lock 201 via the control device 30 (unlocks). In this case as well, the second voice recognition unit 1822 performs voice recognition by referring to the second dictionary D12, so the processing speed of the voice recognition process is fast. The information terminal 10 outputs an unlocking sound from speaker 151 that resembles the unlocking of an electric lock, in order to inform the user that it has received the unlocking command (i.e., that the voice input was successful).

[0133] In the example shown in Figure 4, the information terminal 10 maintains the call state even after the electric lock 201 is unlocked. The user can end the call by touching the termination operation unit 142 (labeled "terminate button" in Figure 4). In other words, when the information terminal 10 receives a touch operation on the termination operation unit 142 while in a call state (a state where it is making a voice call with another terminal), it performs control to disconnect the call with the lobby intercom 20 (end call). The information terminal 10 changes the call screen displayed on the display unit 16 to the standby screen (or home screen) (screen transition). Alternatively, the information terminal 10 may automatically perform control to disconnect the call with the lobby intercom 20 when the electric lock 201 is unlocked.

[0134] Alternatively, the user can also end a call by using voice commands on the microphone 131 while the information terminal 10 is in a call state. As shown in Figure 4, the user speaks the command word "End!" (end command) while in a call state. The end command is a command to end a call with another terminal. When the second voice recognition unit 1822 of the information terminal 10 successfully recognizes the spoken "End!", it performs control to disconnect the call with the lobby intercom 20 (end call). In this case as well, the second voice recognition unit 1822 performs voice recognition by referring to the second dictionary D12, so the processing speed of the voice recognition process is fast. To inform the user that the end command has been received (i.e., that the voice input was successful), the information terminal 10 outputs a "beep" sound and a voice message "Ended" from the speaker 151.

[0135] As such, the control commands for the control processing unit 183 are set to include at least one (in this case, both) of the termination command and the unlock command.

[0136] In Operation Example 1, when the call is ended, the information terminal 10 switches the dictionary used for speech recognition from the second dictionary D12 to the first dictionary D11, which contains command words corresponding to operation events other than call-related operation events. Also, when the call is ended, the information terminal 10 switches the speech recognition method from the second method to the first method. Furthermore, when the call is ended, the information terminal 10 deactivates the second speech recognition unit 1822, etc.

[0137] In other words, the voice recognition unit 182 switches to the first voice recognition mode when it receives a termination operation from the termination operation unit 142 while in a call with another terminal. Furthermore, the voice recognition unit 182 switches to the first voice recognition mode when it successfully recognizes a termination command while in a call with another terminal. Thus, since switching to the first mode is achieved by either a termination operation from the termination operation unit 142 or the utterance of a termination command, user convenience is improved.

[0138] However, the information terminal 10 may automatically perform call disconnection control upon unlocking the electric lock 201. In this case, the voice recognition unit 182 may switch to the first voice recognition mode by receiving an unlocking operation to the unlocking operation unit 143 while in a call with another terminal. Alternatively, the voice recognition unit 182 may switch to the first voice recognition mode by recognizing an unlocking command by voice (successful voice recognition of the unlocking command) while in a call with another terminal. In this way, switching to the first mode is achieved by an unlocking operation to the unlocking operation unit 143 or by uttering an unlocking command, thus improving user convenience.

[0139] In Operation Example 1, the timing for switching from the second dictionary D12 to the first dictionary D11, the timing for switching from the second method to the first method, and the timing for switching the second voice recognition unit 1822, etc., to inactive are all the same. However, some or all of these timings may be different. Also, for example, at least one of these timings may be the same as the timing when unlocking is performed.

[0140] In this operation example 1, the sequence of operations from receiving a call to ending it was explained in two cases: one in which only touch operations are performed on the operation unit 14 (upper part of Figure 4), and another in which only voice operations are performed on the microphone 131 (lower part of Figure 4). However, the information terminal 10 can also be used in a mixed manner, such as answering a call with a touch operation, unlocking with a voice operation, and ending the call with a touch operation, or answering a call with a voice operation, unlocking with a touch operation, and ending the call with a voice operation.

[0141] In this example of operation 1, the speech recognition unit 182 performs speech recognition using the first method when it is in a call state where it is receiving a call from another terminal to the information terminal 10. In other words, the user needs to utter a trigger word, "○○", before uttering a response command in the call state.

[0142] However, the voice recognition unit 182 may perform voice recognition using the second method not only when it is in a call with another terminal, but also when it is in a call state where it is receiving a call from another terminal. For example, the information terminal 10 may switch to a state of voice recognition using the second method when an incoming call occurs. In this case, when answering a call in the call state, the user only needs to speak an answer command without speaking a trigger word.

[0143] When the information terminal 10 is in call mode, it displays the visitor's image captured by the imaging unit 27 of the lobby intercom 20 in real time on the incoming call screen of the display unit 16. The user may see the visitor displayed on the incoming call screen of the display unit 16 and realize that the visitor is a door-to-door salesperson or a stranger, or they may have an urgent matter that prevents them from answering the call and wish to cancel the call. In this case, the information terminal 10 cancels the call while in call mode, for example, by touching the termination operation unit 142 or by voice commanding the termination command. When the information terminal 10 cancels the call, the lobby intercom 20 may automatically notify the visitor by voice or on screen that the call could not be transferred, for example, because the user was out and did not answer, in order to avoid causing distrust.

[0144] Furthermore, if a user sees a visitor displayed on the incoming call screen of the display unit 16 and recognizes that it is a family member or acquaintance, they may want to skip the call and unlock the electric lock 201, inviting the person in before the front door 200. In this case, the information terminal 10, while in the call state, will unlock the electric lock 201 and cancel the call simultaneously, for example, by touching the unlocking operation unit 143 to unlock the electric lock 201, or by voice commanding the unlock command.

[0145] Even if the system had switched to the second voice recognition method at the time an incoming call occurred, the voice recognition unit 182 may switch back to the first voice recognition method when the call is canceled.

[0146] (3-2) Example of operation 2 The following describes an example of the operation of the information terminal 10 (Operation Example 2) regarding the "security set" (see Table 1), which corresponds to operation events other than "call-related" among the intercom control operation events, with reference to Figures 5A and 5B.

[0147] Figure 5A is a conceptual diagram showing a series of operations of the information terminal 10 from standby mode to the execution of the security system, arranged chronologically from left to right along with the screen of the display unit 16.

[0148] Figure 5B is a conceptual diagram showing a series of operations of the information terminal 10 from the state in which the security system is activated (hereinafter referred to as the "alert state") to the state in which the security system is deactivated, arranged chronologically from left to right along with the screen of the display unit 16. The alert state is one of the mode operation states.

[0149] The upper sections of Figures 5A and 5B, similar to the upper section of Figure 4, show the operation of the information terminal 10 when the user requests control and the operation means is a touch operation on the operation unit 14 (or display unit 16). On the other hand, the lower sections of Figures 5A and 5B, similar to the lower section of Figure 4, show the operation of the information terminal 10 when the operation means is a voice operation on the microphone 131.

[0150] First, "Standby" in Figure 5A indicates that the information terminal 10 is in a standby state. In the standby state, the screen of the display unit 16 of the information terminal 10 is, for example, the standby screen.

[0151] While the information terminal 10 is in standby mode, the user can access (launch) the home screen by touching the home control unit (labeled "Home Button" in Figure 5A), which is one of the control units 14. The home screen displays menus for requesting various operation controls. For example, the user can activate the security set by touching the screen area labeled "Security Set" (labeled "Security Button" in Figure 5A) on the home screen. In other words, when the information terminal 10 receives a touch operation requesting the start of the "Security Set" while in standby mode, it switches from standby mode to alert mode, which is one of the operating modes.

[0152] Alternatively, the user can also activate the security system by using voice commands on the microphone 131 while the information terminal 10 is in standby mode. As shown in Figure 5A, the user speaks a trigger word and a command word (control command), "Set the alarm," while in standby mode. When the first voice recognition unit 1821 of the information terminal 10 successfully recognizes the trigger word "Set the alarm," it activates the second voice recognition unit 1822 and other components. Furthermore, when the second voice recognition unit 1822 successfully recognizes the subsequent spoken phrase "Set the alarm," the information terminal 10 switches from standby mode to alert mode. The voice recognition unit 182 maintains the first dictionary D11 for reference while in alert mode, similar to the standby mode, and performs voice recognition. The voice recognition unit 182 also performs voice recognition using the first method, which involves trigger voice recognition and command voice recognition if the trigger word is included, similar to the standby mode. The information terminal 10 outputs a voice message from speaker 151 saying, "Entering alert mode in ** seconds..." to inform the user that it has received the input of the security set control command (i.e., that the voice input was successful).

[0153] While the system is in alert mode, the information terminal 10 displays a screen on the display unit 16 that shows, for example, "Security Pre-Alarm" as shown in Figure 5B. While the system is in alert mode, the information terminal 10 monitors detection signals from sensors installed in the windows of the dwelling unit E2, and triggers an alarm if it detects an intruder entering through the windows. This security system can be set, for example, before the user goes to bed or before going out.

[0154] Subsequently, while in the alerted state, the user can access (launch) the home screen by touching the home control panel (labeled "Home button" in Figure 5B). For example, the user can deactivate the security system by touching the screen area on the home screen that displays "Deactivate Security System" (labeled "Security button" in Figure 5B). In other words, when the information terminal 10 receives a touch request to deactivate the "security system" that is running while in the alerted state, it deactivates the alerted state and returns to its original standby state.

[0155] Alternatively, the user can also deactivate the security set by using voice commands on the microphone 131 while the system is in an alerted state. As shown in Figure 5B, the user speaks a trigger word and a command word (control command) such as "○○, deactivate the security set" while the system is in an alerted state. The second voice recognition unit 1822, etc., may remain active while the system is in an alerted state, or it may temporarily revert to an inactive state even while the system is in an alerted state until it successfully recognizes the next trigger word. When the first voice recognition unit 1821 successfully recognizes the trigger word and the second voice recognition unit 1822 successfully recognizes "deactivate the security set", it switches from the alerted state to a standby state and displays the home screen on the display unit 16. The information terminal 10 outputs a voice message "Security set deactivated" from the speaker 151 to inform the user that it has received input of the control command to deactivate the security set (i.e., the voice input was successful). The information terminal 10 will return to the standby screen if it does not receive any operation input for a certain period of time after the home screen is displayed.

[0156] In this operation example 2, the information terminal 10 performs speech recognition using the first method and the first dictionary D11 without switching the speech recognition method or dictionary.

[0157] In example 2, the information terminal 10 can also handle mixed usage scenarios involving touch and voice operations, such as activating the security set with touch operation and deactivating it with voice operation, or activating the security set with voice operation and deactivating it with touch operation.

[0158] (3-3) Example of operation 3 The following describes an example of the operation of the information terminal 10 (operation example 3) in response to "door monitor confirmation" (see Table 1), which corresponds to operation events other than "call-related" among the intercom control operation events, with reference to Figure 6. Note that "door monitor confirmation" here refers to, for example, checking the real-time video captured by the door monitor (imaging unit) of the entrance station 40 on the screen of the display unit 16 of the information terminal 10.

[0159] Figure 6 is a conceptual diagram showing a series of operations of the information terminal 10 from the standby state to the execution of the door monitor check operation, and from the end of the state in which the door monitor check operation is being performed (hereinafter referred to as the "monitor check state"), arranged chronologically from left to right along with the screen of the display unit 16. The monitor check state is one of the mode operation states.

[0160] The upper part of Figure 6, similar to the upper part of Figure 4, shows the operation of the information terminal 10 when the user requests control and the operating means is a touch operation on the operation unit 14 (or display unit 16). On the other hand, the lower part of Figure 6, similar to the lower part of Figure 4, shows the operation of the information terminal 10 when the operating means is a voice operation on the microphone 131.

[0161] First, "Standby" in Figure 6 indicates that the information terminal 10 is in a standby state. In the standby state, the screen of the display unit 16 of the information terminal 10 is, for example, the standby screen.

[0162] While the information terminal 10 is in standby mode, the user can access (launch) the home screen by touching the home control unit (labeled "home button" in Figure 6), which is one of the control units 14. For example, the user can perform the door monitor confirmation action by touching the image area (play mark in Figure 6) on the home screen, which indicates the start of the door monitor video display. In other words, when the information terminal 10 receives a touch operation requesting the start of "door monitor confirmation" while in standby mode, it switches from standby mode to monitor confirmation mode, which is one of the mode operation states.

[0163] Alternatively, the user can also perform the door monitor confirmation operation by using voice commands on the microphone 131 while the information terminal 10 is in standby mode. As shown in Figure 6, the user speaks the trigger word "○○, show me the door monitor!" and a command word (control command) while in standby mode. When the first voice recognition unit 1821 of the information terminal 10 successfully recognizes the trigger word "○○", it activates the second voice recognition unit 1822 and other components. Furthermore, when the second voice recognition unit 1822 successfully recognizes the subsequent spoken phrase "Show me the door monitor", the information terminal 10 switches from standby mode to monitor confirmation mode. During the monitor confirmation mode, the voice recognition unit 182 maintains the first dictionary D11 for reference and performs voice recognition, similar to the standby mode. Also, during the monitor confirmation mode, the voice recognition unit 182 performs voice recognition using the first method, which involves trigger voice recognition and command voice recognition if the trigger word is included, similar to the standby mode. The information terminal 10 emits a "beep" sound from the speaker 151 to inform the user that it has received the control command input for checking the door monitor (i.e., that the voice input was successful).

[0164] While the information terminal 10 is in monitor confirmation mode (indicated as "Playing" in Figure 6), it displays the real-time video being captured by the door monitor of the entrance station 40 on the display unit 16 (indicated as "Monitor screen" in Figure 6).

[0165] Subsequently, the user can terminate the door monitor check operation by touching the termination operation unit 142 (labeled "termination button" in Figure 6), which can also accept calls ending operations, while in the monitor check state. In other words, when the information terminal 10 receives a touch operation requesting the termination of the "door monitor check" that is currently running in the monitor check state, it terminates the monitor check and returns to its original standby state.

[0166] Alternatively, the user can also terminate the door monitor check operation by using voice commands on the microphone 131 while in the monitor check state. As shown in Figure 6, the user speaks a trigger word and a command word (control command) such as "○○, terminate!" while in the monitor check state. The second voice recognition unit 1822, etc., may remain active while in the monitor check state, or it may return to an inactive state even while in the monitor check state until it successfully recognizes the next trigger word. When the first voice recognition unit 1821 successfully recognizes the trigger word and the second voice recognition unit 1822 successfully recognizes "terminate", the monitor check state is terminated and the system switches to a standby state and displays the standby screen. The information terminal 10 outputs a "beep" sound and a voice message "terminated" from the speaker 151 to inform the user that it has received input of a control command to terminate the door monitor check (i.e., the voice input was successful).

[0167] Thus, in this example of operation 3, the information terminal 10 performs speech recognition using the first method and the first dictionary D11 without switching the speech recognition method or dictionary.

[0168] In example 3, the information terminal 10 can also handle mixed usage scenarios involving touch and voice operations, such as performing a door monitor check operation with touch operation and ending the operation with voice operation, or performing a door monitor check operation with voice operation and ending the operation with touch operation.

[0169] (3-4) Example of operation 4 The following describes an example of the operation of the information terminal 10 (operation example 4) in relation to "recorded video playback" (see Table 1), which corresponds to operation events other than "call-related" events among the intercom control operation events, with reference to Figure 7. Note that recorded video playback here refers to, for example, the playback of a video recording captured by the imaging unit 27 of a visitor who called the information terminal 10 via the lobby intercom 20, and a recorded message from the visitor conveying their request.

[0170] Figure 7 is a conceptual diagram showing a series of operations of the information terminal 10, from the standby state to the execution of recording and playback operations, and from the state in which recording and playback operations are being performed (hereinafter referred to as the "recording and playback state"), arranged chronologically from left to right along with the screen of the display unit 16. The recording and playback state is one of the mode operation states.

[0171] The upper part of Figure 7, similar to the upper part of Figure 4, shows the operation of the information terminal 10 when the user requests control and the operating means is a touch operation on the operation unit 14 (or display unit 16). On the other hand, the lower part of Figure 7, similar to the lower part of Figure 4, shows the operation of the information terminal 10 when the operating means is a voice operation on the microphone 131.

[0172] First, "Standby" in Figure 7 indicates that the information terminal 10 is in a standby state. In the standby state, the screen of the display unit 16 of the information terminal 10 is, for example, the standby screen.

[0173] While the information terminal 10 is in standby mode, the user can access (launch) the home screen by touching the home control unit (labeled "home button" in Figure 7), which is one of the control units 14. For example, the user can initiate recording playback by touching the image area (playback mark in Figure 7) on the home screen, which indicates the start of recording playback. In other words, when the information terminal 10 receives a touch operation requesting the start of "recording playback" while in standby mode, it switches from standby mode to the recording playback state, which is one of the mode operation states.

[0174] Alternatively, the user can also perform recording and playback operations by using voice commands on the microphone 131 while the information terminal 10 is in standby mode. As shown in Figure 7, the user speaks a trigger word and a command word (control command), "○○, record and play back," while in standby mode. When the first voice recognition unit 1821 of the information terminal 10 successfully recognizes the trigger word "○○," it activates the second voice recognition unit 1822 and other components. Furthermore, when the second voice recognition unit 1822 successfully recognizes the subsequent spoken command "record and play back," the information terminal 10 switches from standby mode to recording and playback mode. During recording and playback mode, the voice recognition unit 182 maintains the first dictionary D11 for reference and performs voice recognition, similar to the standby mode. Also, during recording and playback mode, the voice recognition unit 182 performs voice recognition using the first method, which involves trigger voice recognition and command voice recognition if the trigger word is included, similar to the standby mode. The information terminal 10 outputs the voice message "There is unread information. Playing the recording." from speaker 151 to inform the user that it has received input for the control command to play the recording (i.e., that the voice input was successful).

[0175] During the recording playback state (indicated as "Playback" in Figure 7), the information terminal 10 outputs the recorded video and audio of the visitor from the display unit 16 and speaker 151 (indicated as "Playback screen" in Figure 7).

[0176] Subsequently, the user can terminate the recording playback operation by touching the termination operation unit 142 (labeled "termination button" in Figure 7), which can also accept calls ending operations, when playback is finished (or during playback) while in recording playback mode. In other words, when the information terminal 10 receives a touch operation requesting the termination of the "recording playback" currently running in recording playback mode, it terminates the recording playback mode and returns to its original standby state.

[0177] Alternatively, the user can also terminate the recording playback operation by using voice commands on the microphone 131 while in recording playback mode. As shown in Figure 7, the user speaks a trigger word and a command word (control command) such as "○○, stop!" while in recording playback mode. The second voice recognition unit 1822, etc., may remain active during recording playback mode, or it may temporarily revert to an inactive state even during recording playback mode until it successfully recognizes the next trigger word. When the first voice recognition unit 1821 successfully recognizes the trigger word and the second voice recognition unit 1822 successfully recognizes "stop", it terminates the recording playback mode and switches to standby mode, displaying the standby screen. The information terminal 10 outputs a "beep" sound and a voice message "Finished" from the speaker 151 to inform the user that it has received input of the control command to terminate recording playback (i.e., the voice input was successful).

[0178] Thus, in this example of operation 4, the information terminal 10 performs speech recognition using the first method and the first dictionary D11 without switching the speech recognition method or dictionary.

[0179] In example 4, the information terminal 10 can also handle mixed usage scenarios involving touch and voice operations, such as performing recording and playback operations with touch operation and ending the operation with voice operation, or performing recording and playback operations with voice operation and ending the operation with touch operation.

[0180] (3-5) Example of operation 5 The following describes an example of the operation of the information terminal 10 (Operation Example 5) related to the "air conditioner" (see Table 1), which corresponds to equipment control operation events rather than intercom control, with reference to Figure 8. Note that the air conditioner referred to here is one of several devices installed in dwelling unit E2 that can be controlled via the HEMS controller.

[0181] Figure 8 is a conceptual diagram showing a series of operations of the information terminal 10 from the standby state to the execution of an air conditioner control operation (for example, the operation of turning on the air conditioner and starting it up), arranged chronologically from left to right along with the screen of the display unit 16. Note that in operation example 5, the explanation of the control operation to stop the air conditioner (stop operation) is omitted.

[0182] The upper part of Figure 8, similar to the upper part of Figure 4, shows the operation of the information terminal 10 when the user requests control and the operating means is a touch operation on the operation unit 14 (or display unit 16). On the other hand, the lower part of Figure 8, similar to the lower part of Figure 4, shows the operation of the information terminal 10 when the operating means is a voice operation on the microphone 131.

[0183] First, "Standby" in Figure 8 indicates that the information terminal 10 is in a standby state. In the standby state, the screen of the display unit 16 of the information terminal 10 is, for example, the standby screen.

[0184] While the information terminal 10 is in standby mode, the user can access (launch) the home screen by touching the home control unit (labeled "Home Button" in Figure 8), which is one of the control units 14. For example, the user touches an image area on the home screen that indicates the start of a device control operation (labeled "Operate" in Figure 8). The display unit 16 then displays a screen for selecting the device to be controlled (labeled "Device Control Screen" in Figure 8). The user can then execute the air conditioner control operation by touching an image area labeled "Air Conditioner" on the device control screen. In other words, when the information terminal 10 receives a touch operation requesting the start of the "Air Conditioner Control Operation" while in standby mode, it executes the operation for air conditioner control. Specifically, the information terminal 10 sends an instruction signal to the HEMS controller to operate the specified air conditioner. When the HEMS controller receives the instruction signal, it controls the specified air conditioner to operate (on control).

[0185] Alternatively, the user can also perform air conditioner control operations by using voice commands on the microphone 131 while the information terminal 10 is in standby mode. As shown in Figure 8, the user speaks a trigger word and a command word (control command), such as "○○, turn on the air conditioner," while in standby mode. When the first voice recognition unit 1821 of the information terminal 10 successfully recognizes the trigger word "○○," it activates the second voice recognition unit 1822 and other components. Furthermore, when the second voice recognition unit 1822 successfully recognizes the subsequent spoken command "Turn on the air conditioner," the information terminal 10 performs air conditioner control operations and returns to standby mode. The voice recognition unit 182 maintains the first dictionary D11 as the reference dictionary for air conditioner control operations as well. The voice recognition unit 182 also performs voice recognition for air conditioner control operations using the first method, which involves trigger voice recognition and command voice recognition if the trigger word is included. Although not shown in Figure 8, it is preferable that the information terminal 10 outputs a "beep" sound from the speaker 151 to inform the user that it has received input of a control command for the operation of the air conditioner (i.e., that the voice input was successful).

[0186] Thus, in example 5, the information terminal 10 performs speech recognition using the first method and the first dictionary D11 without switching the speech recognition method or dictionary.

[0187] (3-6) Example of operation 6 The following describes an example of the operation of the information terminal 10 (Operation Example 6) for the "Return Home Operation" event of scene control, rather than intercom control, with reference to Figure 9. Here, it is assumed that the command word "I'm home" corresponding to the return home operation event is associated with the execution of the "record and play back" operation described in Operation Example 4. In other words, when the user utters the control command "I'm home," the information terminal 10 executes the "record and play back" operation described in Operation Example 4. That is, when the information terminal 10 receives a touch operation requesting the start of scene control execution while in standby mode, it switches from standby mode to the recording and play back state, which is one of the mode operation states.

[0188] While a detailed explanation is omitted here, each command word for scene control shown in Table 1 corresponds to a specific operation of the information terminal 10, and this is set along with the scene number and stored in the information terminal 10's memory unit 17. The user can change the settings related to scene control as appropriate on the settings screen of the information terminal 10's display unit 16.

[0189] Figure 9 is a conceptual diagram showing a series of operations of the information terminal 10 from the standby state to the execution of a scene control operation (in this case, a recording and playback operation), arranged chronologically from left to right along with the screen of the display unit 16.

[0190] The upper part of Figure 9, similar to the upper part of Figure 4, shows the operation of the information terminal 10 when the user requests control and the operation means is a touch operation on the operation unit 14 (or display unit 16). Note that the upper part of Figure 9 is the same as the flow from "standby" to "playback" in the upper part of Figure 7, which was referenced in the explanation of operation example 4. Therefore, the explanation of the execution of the recording playback operation by touch operation on the operation unit 14 (or display unit 16) is omitted here.

[0191] The lower part of Figure 9, similar to the lower part of Figure 4, shows the operation of the information terminal 10 when the operating means is voice operation to the microphone 131.

[0192] First, "Standby" in Figure 9 indicates that the information terminal 10 is in a standby state. In the standby state, the screen of the display unit 16 of the information terminal 10 is, for example, the standby screen.

[0193] The user can also execute scene control operations by using voice commands on the microphone 131 while the information terminal 10 is in standby mode. As shown in Figure 9, the user speaks the trigger word "I'm home, XX!" and the command word (control command) while in standby mode. When the first voice recognition unit 1821 of the information terminal 10 successfully recognizes the trigger word "XX", it activates the second voice recognition unit 1822, etc. Furthermore, when the second voice recognition unit 1822 successfully recognizes the next spoken word "I'm home", the information terminal 10 executes the recording playback operation as the scene control associated with "I'm home" (the operation event upon returning home). To inform the user that it has received input for the scene control command (i.e., that the voice input was successful), the information terminal 10 outputs the voice message "Welcome back. There is unread information. Playing the recording." from the speaker 151. In the example shown in Figure 9, the information terminal 10 outputs a message from speaker 151 asking, "Do you want to set a scene* (scene control number)?". The user responds by saying "Yes" or "No" (see "Other" in Table 1), allowing the information terminal 10 to accept the change in the scene control number setting. In short, the information terminal 10 can accept changes to various setting information interactively by outputting a query message to the user at a predetermined timing (or at any timing after receiving an operation).

[0194] Furthermore, the command word "Tadaima" (I'm home), which corresponds to the operation event upon returning home, can be associated not only with recording and playback, but also with actions such as turning on lighting equipment (on control) or operating air conditioners (on control). In this case, the information terminal 10 sends an instruction signal to the HMES controller to also perform these on-control actions as a scene control associated with "Tadaima". The control results for the lighting equipment and air conditioner are not displayed on the screen of the display unit 16, but they may be displayed.

[0195] Thus, in this example of operation 6, the information terminal 10 performs speech recognition using the first method and the first dictionary D11 without switching the speech recognition method or dictionary.

[0196] (4) Advantages of the first and second methods As described above, in the information terminal 10 of this embodiment, when making a call with another terminal, speech recognition is performed using a second method, which is simpler than the first method. Therefore, the processing speed related to speech recognition during a call is likely to improve, potentially enhancing user convenience. As a result, the information terminal 10 has the advantage of being able to improve its speech recognition function.

[0197] Furthermore, since voice recognition is performed using a second method that does not involve trigger voice recognition during calls with other devices, users do not need to utter trigger words during calls, which can reduce user stress. As a result, user convenience may be improved. In addition, there are concerns about reduced security if trigger words become known to anyone other than the user, but with information terminal 10, trigger words are not required during calls with other devices. Therefore, the possibility of trigger words uttered by the user during a call being heard by the other party is reduced. As a result, a decrease in security can also be mitigated. Moreover, if a user suddenly utters a trigger word during a call, it can feel strange to the other party, but with information terminal 10, trigger words are not required during calls with other devices, so the occurrence of such strangeness can also be suppressed.

[0198] Furthermore, in the information terminal 10 of this embodiment, a second method, which is simpler than the first method of speech recognition, is implemented by the difference in the number of control commands registered in the dictionary, thus improving the processing speed of speech recognition during a call with a simpler configuration.

[0199] (5) Control commands and operating status As can be understood from the above operation examples 1 to 6 and Table 1, in this embodiment, the types of control commands recognized by the speech recognition unit 182 are set according to the operating state of the device (information terminal 10). Therefore, in each operating state of the device (information terminal 10), when the speech recognition unit 182 identifies which control commands are included in the voice input to the microphone 131, it is possible to reduce the possibility of searching by referencing all control commands. Thus, the processing speed related to speech recognition can be improved.

[0200] Furthermore, if all control commands are included in the search criteria regardless of the operating state, the recognition accuracy for speech recognition may decrease. For example, the possibility of the speech recognition unit 182 failing to recognize speech may increase even if the user is speaking the correct command word. However, in the information terminal 10 of this embodiment, such a decrease in recognition accuracy can be suppressed.

[0201] Thus, the information terminal 10 can improve the processing speed or accuracy of speech recognition. As a result, the information terminal 10 has the advantage of being able to improve the functionality of speech recognition.

[0202] Specifically, when the operating state is a calling state where the device is receiving a call from another terminal, the types of control commands recognized by voice include response commands for answering a call. However, when the operating state is a standby state, the types of control commands recognized by voice do not include response commands. Therefore, in the standby state, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131, response commands for calls are not set as the search target. As a result, the processing speed or accuracy of voice recognition may be improved. In this embodiment, by switching between the first dictionary D11 and the second dictionary D12 used depending on the operating state, response commands are not included as the search target when the device is in a standby state.

[0203] Furthermore, when the operating state is a calling state where the device is receiving a call from another terminal, the types of control commands recognized by voice include the unlock command for unlocking the electric lock 201. However, when the operating state is a standby state, the types of control commands recognized by voice do not include the unlock command. Therefore, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131 for the standby state, the unlock command for the electric lock 201 is not set as a search target. As a result, the processing speed or accuracy of voice recognition may be improved. In this embodiment, by switching between the first dictionary D11 and the second dictionary D12 used depending on the operating state, the unlock command is not included as a search target when the device is in a standby state.

[0204] Furthermore, when the operating state is a calling state where a call is being received from another terminal, the types of control commands recognized by voice include a cancel command to cancel a call. When the operating state is a standby state, the types of control commands recognized by voice do not include a cancel command. Therefore, for the standby state, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131, the call cancel command is not set as a search target. As a result, the processing speed or accuracy of voice recognition may be improved. The "cancel command" for a call referred to here may correspond to, for example, a call termination command or an unlock command to unlock the electric lock 201. Alternatively, it may be a dedicated command for canceling a call (for example, "Stop calling"). In this embodiment, by switching between the first dictionary D11 and the second dictionary D12 used depending on the operating state, the call cancel command is not searched when the device is in a standby state.

[0205] Furthermore, when the operating state is in a call state with another terminal, the types of control commands recognized by voice include the unlock command for unlocking the electric lock 201. However, when the operating state is in standby mode, the types of control commands recognized by voice do not include the unlock command. Therefore, in standby mode, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131, the unlock command for the electric lock 201 is not set as a search target. As a result, the processing speed or accuracy of voice recognition may be improved. In this embodiment, by switching between the first dictionary D11 and the second dictionary D12 used depending on the operating state, the unlock command is not included as a search target when in standby mode.

[0206] Furthermore, when the operating state is a call with another terminal, the types of control commands recognized by voice include the call termination command. When the operating state is a standby state, the types of control commands recognized by voice do not include the call termination command. Therefore, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131 in the standby state, the call termination command is not set as a search target. As a result, the processing speed or accuracy of voice recognition may be improved. In this embodiment, by switching between the first dictionary D11 and the second dictionary D12 used depending on the operating state, the call termination command is not included as a search target when the device is in a standby state.

[0207] Furthermore, when the operating state is executing a predetermined operating mode, the types of control commands recognized by voice do not include start commands to initiate the predetermined operating mode, but include release commands to cancel the predetermined operating mode that is currently running. The predetermined operating modes are, as described above, the operating mode for the security set, the operating mode for checking the door monitor, and the operating mode for recording and playback. These start commands are, in Table 1, "set the alarm," "show me the door monitor," and "play back the recording," respectively. Therefore, for operating states that are executing a predetermined operating mode, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131, the start commands for the predetermined operating mode are not set as search targets. For example, for an alarm state where the security set is already in operation, when the voice recognition unit 182 identifies which control commands are included in the input voice, the start command for the security set (set the alarm) is not set as a search target. As a result, the processing speed or accuracy of voice recognition may be improved. In this embodiment, the first dictionary D11 is configured so that when a predetermined operating mode is in operation, its start command is not included as a search target.

[0208] Furthermore, when the operating state is in standby mode, the types of control commands recognized by voice do not include the release commands that cancel the predetermined operating modes described above, but include the start commands that initiate the predetermined operating modes. The predetermined operating modes are, as described above, the operating modes for the security set, the door monitor confirmation, and the recording and playback. The release commands for these modes are, in Table 1, "cancel the security set," "end," and "end," respectively. Therefore, when the voice recognition unit 182 identifies which control commands are included in the voice input to the microphone 131 in standby mode, the release commands for the predetermined operating modes are not set as search targets. As a result, the processing speed or accuracy of voice recognition may be improved. In this embodiment, in the first dictionary D11, when the system is in standby mode, the release commands that cancel the predetermined operating modes are not included as search targets.

[0209] (6) Variant (6-1) Torture 1 The information terminal 10 and intercom system 1 according to this modified example (Modification Example 1) will be described below with reference to Figure 10. Note that, for the information terminal 10 and intercom system 1 according to Modification Example 1, components that are substantially common with the information terminal 10 and intercom system 1 according to the above embodiment may be given the same reference numerals, and their detailed descriptions may be omitted.

[0210] The intercom system 1 according to Modification 1 differs from the intercom system 1 according to the above embodiment in that it further comprises a voice recognition system A1. The voice recognition system A1 includes, for example, a microcomputer having a processor and memory. The microcomputer functions as the voice recognition system A1 by executing a program stored in memory, which is performed by the processor. The program executed by the processor is pre-recorded in the memory of the microcomputer in this example, but it may also be recorded on a non-temporary recording medium such as a memory card and provided, or it may be provided via a telecommunication line such as the Internet.

[0211] In Modification 1, as shown in Figure 10, the functions of the speech recognition system A1 are provided on an external server 70. The server 70 consists of one or more server devices, and such server devices can constitute a cloud system.

[0212] For convenience, Figure 10 shows only the information terminal 10 installed in a certain dwelling unit E2 of the intercom system 1, and the server 70 to which the information terminal 10 is connected via network NT1.

[0213] As shown in Figure 10, the information terminal 10 according to Modification 1 comprises a first communication unit 11, a second communication unit 12, a sound acquisition unit 13, an operation unit 14, a sound output unit 15, a display unit 16, a storage unit 17, a processing unit 18, and a third communication unit 19. The processing unit 18 comprises a voice processing unit 181, a voice recognition unit 182, a control processing unit 183, a display processing unit 184, and a call processing unit 185. The voice recognition unit 182 performs voice recognition based on the sound input to the microphone 131. The first communication unit 11 performs communication, including calls, with other terminals (in this case, the lobby intercom 20).

[0214] The second communication unit 12 communicates with the voice recognition system A1. Second Communications Department 12 The control processing unit 183 communicates with a voice recognition system A1 that performs voice recognition of control commands related to the control performed by the control processing unit 183. The third communication unit 19 performs communication, including calls, with other terminals (in this case, the door station 40).

[0215] The information terminal 10 according to Modification 1 has a first mode and a second mode. In the first mode, the voice recognition unit 182 performs trigger voice recognition to determine whether or not a trigger word is included in the voice input to the microphone 131. In the first mode, if a trigger word is included, it communicates with the voice recognition system A1 to perform command voice recognition. In the second mode, the voice recognition unit 182 does not perform trigger voice recognition, but communicates with the voice recognition system A1 to perform command voice recognition. The information terminal 10 operates in the first mode when in standby mode and in the second mode when in call mode with another terminal.

[0216] As shown in Figure 10, the speech recognition system A1 comprises a communication unit A11, a speech recognition unit A12, and a storage unit A13. The communication unit A11 has a communication interface for individually communicating with each of the multiple information terminals 10 (or their second communication units 12) installed in one or more facilities (such as an apartment building 5). When the speech recognition unit A12 receives voice data that may include control commands (command words) from the information terminal 10 via the communication unit A11, it performs command speech recognition processing. The speech recognition system A1 returns the speech recognition result from the speech recognition unit A12 to the information terminal 10, which is the source of the voice data, via the communication unit A11.

[0217] The storage unit A13 stores the command dictionary D1 (first dictionary D11 and second dictionary D12) in place of the storage unit 17 of the information terminal 10 according to the above embodiment. The storage unit 17 of the information terminal 10 stores the trigger dictionary D0, as in the above embodiment.

[0218] In short, the voice recognition unit 182 of the information terminal 10 in the modified example 1 has only the trigger voice recognition function of the information terminal 10 in the above embodiment, and the command voice recognition function of the information terminal 10 in the above embodiment is provided in an external system (voice recognition system A1) of the information terminal 10.

[0219] In the modified example 1, the information terminal 10 performs trigger voice recognition locally and has the voice recognition system A1 perform command voice recognition. Therefore, the processing load on the information terminal 10 can be reduced.

[0220] Furthermore, the information terminal 10 according to Modification 1, like the information terminal 10 according to the above embodiment, does not perform trigger voice recognition during a call, which increases the likelihood of improved processing speed related to voice recognition and potentially enhances user convenience. It also helps to suppress a decrease in security. As a result, the information terminal 10 according to Modification 1 also has the advantage of being able to improve the voice recognition function.

[0221] In Modification 1, the speech recognition system A1 is highly likely to constantly monitor whether the voice input to the information terminal 10 contains a command word. As a result, the amount of communication related to voice data between the information terminal 10 and the speech recognition system A1 may increase. Considering this point, in Modification 1, it is preferable to limit the timing of command voice recognition and the types of voices that are subject to command voice recognition. Specifically, it is preferable that the information terminal 10 is configured to be able to perform command voice recognition in a certain state (for example, a standby state).

[0222] The same functions as those of the information terminal 10 according to Modification 1 can also be realized in the control method. The control method includes a speech recognition step, a control processing step, a first communication step, and a second communication step. In the speech recognition step, speech recognition processing is performed based on the voice input to the microphone 131. In the first communication step, communication is performed, including a call with another terminal. In the second communication step, communication is performed with a speech recognition system that performs command speech recognition for control commands related to the control performed in the control processing step. The information terminal 10 has a first mode and a second mode. In the first mode, trigger speech recognition is performed in the speech recognition step to determine whether or not a trigger word is included in the voice input to the microphone 131. Also in the first mode, if a trigger word is included, communication is performed with the speech recognition system A1 in the second communication step to have command speech recognition performed. In the second mode, trigger speech recognition is not performed in the speech recognition step, and communication is performed with the speech recognition system A1 in the second communication step to have command speech recognition performed. The control method controls the information terminal 10 to operate in a first mode when in standby mode and in a second mode when in a call state with another terminal. This control method also has the advantage of being able to improve the voice recognition function. This control method is used on a computer system (information terminal 10). In other words, this control method can also be implemented in a program. The program relating to Modification 1 is a program that causes one or more processors to execute the above control method. The program may be recorded on a non-temporary recording medium that can be read by a computer.

[0223] (6-2) Modification 2 The following description of the modified version (Modification 2) of the information terminal 10 and intercom system 1 will be made with reference to Figure 11. Note that for the information terminal 10 and intercom system 1 of Modification 2, components that are substantially common with the information terminal 10 and intercom system 1 of the above embodiment and Modification 1 may be given the same reference numerals, and their detailed descriptions may be omitted.

[0224] Modification 2 is a further modification of Modification 1 described above. The intercom system 1 according to Modification 2 also further includes a voice recognition system A1, similar to Modification 1 described above. However, the voice recognition system A1 according to Modification 2 differs from the voice recognition system A1 according to Modification 1 described above in that it has not only a command voice recognition function but also a trigger voice recognition function. In other words, the information terminal 10 according to Modification 2 differs from the information terminal 10 according to Modification 1 described above in that it does not have either voice recognition function.

[0225] For convenience, Figure 11 shows only the information terminal 10 installed in dwelling unit E2 of the intercom system 1, and the server 70 to which the information terminal 10 is connected via network NT1.

[0226] The voice recognition system A1 according to the modified example 2 comprises a communication unit A11 that communicates with an information terminal 10, a voice recognition unit A12, and a storage unit A13.

[0227] When the voice recognition unit A12 receives the voice input to the microphone 131 from the information terminal 10 via the communication unit A11, it performs voice recognition on the voice. When the voice recognition unit A12 receives voice data from the information terminal 10 via the communication unit A11 that may include trigger words and / or control commands (command words), it may perform trigger voice recognition and / or command voice recognition processing. In other words, the voice recognition unit A12 performs command voice recognition for control commands related to the control performed by the control processing unit 183 of the information terminal 10. The voice recognition system A1 sends the voice recognition result from the voice recognition unit A12 back to the information terminal 10, the source of the voice data, via the communication unit A11.

[0228] Memory unit A13 stores the trigger dictionary D0 and the command dictionary D1.

[0229] Here, the voice recognition unit A12 according to the modified example 2 has a first mode and a second mode. In the first mode, trigger voice recognition is performed to determine whether or not a trigger word is included in the voice input to the microphone 131. Also in the first mode, if a trigger word is included, command voice recognition is performed for control commands related to the control performed by the control processing unit 183 of the information terminal 10. In the second mode, command voice recognition is performed without trigger voice recognition. The voice recognition system A1 operates in the first mode when the information terminal 10 is in standby mode, and operates in the second mode when the information terminal 10 is in a call state with other terminals (lobby intercom 20, door station 40). In other words, the voice recognition system A1 receives not only voice data from the information terminal 10, but also information about the operating state of the information terminal 10 at the time the voice input was received (currently). For example, when voice input is received when the information terminal 10 is in a call state, it transmits the user's voice data and information indicating that it is in a call state to the voice recognition system A1. The voice recognition system A1 determines whether to operate in the first mode or the second mode based on the operating status of the received information terminal 10. The type of control command recognized by the voice recognition unit A12 is set according to the operating status of the information terminal 10.

[0230] In the speech recognition unit A12 according to the modified example 2, the part (circuit or server device) that performs the trigger speech recognition process and the part (circuit or server device) that performs the command speech recognition process may be separated.

[0231] On the other hand, the information terminal 10 according to Modification 2, as shown in Figure 11, does not have a voice recognition unit 182, unlike the information terminal 10 according to the above embodiment and Modification 1. Also, the storage unit 17 of the information terminal 10 does not store the trigger dictionary D0 and the command dictionary D1.

[0232] As shown in Modification Example 2, by performing all trigger speech recognition and command speech recognition processing on the speech recognition system A1 (for example, the cloud), the processing load on the information terminal 10 can be further reduced.

[0233] Furthermore, the voice recognition system A1 according to Modification 2, like the information terminal 10 according to the above embodiment, does not perform trigger voice recognition when the information terminal 10 is in a call state, which increases the likelihood of improved processing speed related to voice recognition and can improve user convenience. It can also suppress a decrease in security. As a result, the voice recognition system A1 according to Modification 2 also has the advantage of being able to improve the voice recognition function.

[0234] Furthermore, in the speech recognition system A1 according to Modification 2, the types of control commands recognized by the speech recognition unit A12 are set according to the operating state of the information terminal 10. Therefore, in each operating state, when the speech recognition unit A12 identifies which control commands are included in the voice input to the microphone 131, the possibility of searching by referencing all control commands can be reduced. As a result, the processing speed or accuracy of speech recognition may be improved.

[0235] Furthermore, in variation 2, the processing load on the information terminal 10 can be reduced compared to the case where the information terminal 10 is equipped with a voice recognition function. As a result, the voice recognition system A1 has the advantage of being able to improve its voice recognition function.

[0236] In Modification 2, the speech recognition system A1 is more likely to constantly monitor whether the voice input to the information terminal 10 contains a trigger word or a command word. As a result, Modification 2 may result in an even greater increase in the amount of communication related to voice data than Modification 1. Considering this point, even in Modification 2, it is preferable to narrow down the timing of trigger voice recognition and command voice recognition, and the types of voices that are targeted for voice recognition.

[0237] Furthermore, the same functions as those of the voice recognition system A1 in Modification 2 can also be realized in the control method. One control method for controlling the voice recognition system A1 includes a communication step of communicating with the information terminal 10 and a voice recognition step. In the voice recognition step, when the voice input to the microphone 131 is received from the information terminal 10 in the communication step, voice recognition processing is performed on the voice. The voice recognition system A1 has a first mode and a second mode. In the first mode, in the voice recognition step, trigger voice recognition is performed to determine whether or not the voice input to the microphone 131 contains a trigger word. Also in the first mode, if a trigger word is contained, command voice recognition is performed for control commands related to the control performed by the control processing unit 183 of the information terminal 10. In the second mode, command voice recognition is performed in the voice recognition step without performing trigger voice recognition. In the control method, the system operates in the first mode when the information terminal 10 is in a standby state, and operates in the second mode when the information terminal 10 is in a call state with other terminals (lobby intercom 20, door station 40). The control method described above also has the advantage of being able to improve the functionality of speech recognition. This control method is used on a computer system (speech recognition system A1). In other words, this control method can also be implemented in a program. The program relating to Modification 2 is a program that causes one or more processors to execute the control method described above. The program may be recorded on a non-temporary recording medium that can be read by a computer.

[0238] Another control method for controlling the speech recognition system A1 includes a communication step for communicating with the information terminal 10 and a speech recognition step. In the speech recognition step, when the voice input to the microphone 131 is received from the information terminal 10 in the communication step, speech recognition processing is performed on the voice. In the speech recognition step, command speech recognition is performed on control commands related to the control performed by the control processing unit 183 of the information terminal 10. The type of control command to be speech-recognized in the speech recognition step is set according to the operating state of the information terminal 10. The above alternative control method also has the advantage of allowing improvements to the speech recognition function. This alternative control method is used on a computer system (speech recognition system A1). In other words, this alternative control method can also be implemented in a program. The program relating to Modification 2 is a program for causing one or more processors to execute the above alternative control method. The program may be recorded on a non-temporary recording medium that can be read by a computer.

[0239] (6-3) Other modifications The information terminal 10, intercom system 1, and voice recognition system A1 in this disclosure include a computer system. The computer system mainly consists of a processor and memory as hardware. The functions of the information terminal 10, intercom system 1, and voice recognition system A1 in this disclosure are realized by the processor executing a program recorded in the memory of the computer system. The program may be pre-recorded in the memory of the computer system, provided via a telecommunications line, or provided on a non-temporary recording medium such as a memory card, optical disk, or hard disk drive that can be read by the computer system. The processor of the computer system consists of one or more electronic circuits including semiconductor integrated circuits (ICs) or large-scale integrated circuits (LSIs). The integrated circuits such as ICs or LSIs referred to here are named differently depending on the degree of integration, and include integrated circuits called system LSIs, VLSIs (Very Large Scale Integration), or ULSIs (Ultra Large Scale Integration). Furthermore, FPGAs (Field-Programmable Gate Arrays) that are programmed after the LSI is manufactured, or logic devices capable of reconfiguring the junction relationships or circuit compartments within the LSI, can also be used as processors. Multiple electronic circuits may be integrated onto a single chip or distributed across multiple chips. Multiple chips may be integrated into a single device or distributed across multiple devices. The computer system referred to here includes a microcontroller having one or more processors and one or more memories. Therefore, the microcontroller also consists of one or more electronic circuits, including semiconductor integrated circuits or large-scale integrated circuits.

[0240] Furthermore, it is not essential that the multiple functions of each of the information terminal 10, the intercom system 1, and the voice recognition system A1 be integrated into a single housing. The components of each of the information terminal 10, the intercom system 1, and the voice recognition system A1 may be distributed across multiple housings. Conversely, the multiple functions of each of the information terminal 10, the intercom system 1, and the voice recognition system A1 may be integrated into a single housing. Moreover, at least some of the functions of each of the information terminal 10, the intercom system 1, and the voice recognition system A1, for example, some of the functions of the information terminal 10, may be implemented by the cloud (cloud computing), etc.

[0241] In the above embodiment, as an example, it was assumed that the information terminal 10 was a dwelling unit terminal (intercom master unit), and that voice recognition was performed based on voice input directly into the microphone 131 equipped on the unit. However, the microphone for acquiring voice for voice recognition may be external to the information terminal 10. Alternatively, the microphone for acquiring voice for voice recognition may be installed in a device other than the information terminal 10 (for example, an intercom sub-unit located in the same dwelling unit E2, or a portable device that the user can carry, such as a smartphone, tablet, or wearable device). The information terminal 10 may communicate with the other device to acquire the user's voice.

[0242] For example, the information terminal 10 or the voice recognition system A1 may perform voice recognition based on the voice input into the microphone of a smartphone carried by the user, which is an example of a mobile terminal. In this case, the smartphone is pre-installed with dedicated application software for communicating with the information terminal 10. It is preferable that the user can use the smartphone to respond to calls to the information terminal 10 or to talk to visitors, even if they are not physically near the information terminal 10, as long as they are in the dwelling unit E2. The smartphone transmits the voice (data) input into the microphone to the information terminal 10 via wireless communication compliant with standards such as Wi-Fi (registered trademark). In other words, if the user is in the dwelling unit E2 and the smartphone is within a certain range of the information terminal 10, the user may use the smartphone to indirectly respond to calls from the information terminal 10 or to talk to them using voice control. Furthermore, at least a part of the voice recognition function of the information terminal 10 may be installed on the smartphone as application software. For example, the trigger voice recognition function may be on the smartphone, and the command voice recognition function may be on the information terminal 10 or the voice recognition system A1.

[0243] In one modified example, if the unlock button 143 is pressed or a command word (e.g., "unlock") is successfully recognized while a call is being made from or while communicating with the entrance unit 40, the information terminal 10 may send an unlock signal to unlock the electric lock on the entrance door of the facility (dwelling unit E2). The information terminal 10 may also unlock the electric lock on the entrance door via the entrance unit 40 or another device.

[0244] In one modified example, if the intercom system 1 is applied to a detached house, when the unlock button 143 is pressed or a command word (e.g., "unlock") is successfully recognized during a call from the door station of the detached house or during a conversation with the door station, the information terminal 10 may send an unlock signal to unlock the electric lock on the front door of the detached house. The information terminal 10 may also unlock the electric lock on the front door via the door station or another device.

[0245] The embodiments, modification 1, modification 2, and other modifications described above can be applied in appropriate combinations.

[0246] (7) Summary As described above, the information terminal (10) according to the first embodiment comprises a voice recognition unit (182), a control processing unit (183), and a communication unit (first communication unit 11, third communication unit 19). The voice recognition unit (182) performs voice recognition based on the voice input to the microphone (131). The control processing unit (183) performs control based on the voice recognition result of the voice recognition unit (182). The communication unit (first communication unit 11, third communication unit 19) performs communication, including calls with other terminals (lobby intercom 20, door station 40). The voice recognition unit (182) performs voice recognition using the first method or a second method which is simpler than the first method of voice recognition. The voice recognition unit (182) performs voice recognition using the first method in standby mode and uses the second method when in a call state with another terminal.

[0247] According to the above configuration, during a call with another terminal, speech recognition is performed using a second method, which is simpler than the first method. Therefore, the processing speed for speech recognition during a call is likely to improve, potentially enhancing user convenience. As a result, the information terminal (10) has the advantage of being able to improve its speech recognition functionality.

[0248] With respect to the information terminal (10) according to the second embodiment, in the first embodiment, the first method performs trigger speech recognition to determine whether or not a trigger word is included in the voice input to the microphone (131). If a trigger word is included, the first method performs command speech recognition for control commands related to the control performed by the control processing unit (183). The second method performs command speech recognition without performing trigger speech recognition on the voice input to the microphone (131).

[0249] According to the above configuration, since speech recognition is performed using a second method that does not involve trigger speech recognition during a call with another device, the user does not need to utter a trigger word during the call. As a result, user convenience may be improved. Furthermore, the possibility of the trigger word uttered by the user being heard by the other party during a call with another device is reduced. As a result, a decrease in security may also be mitigated.

[0250] With respect to the information terminal (10) according to the third embodiment, in the first or second embodiment, the first method performs first command voice recognition for first control commands relating to predetermined controls performed by the control processing unit (183) in standby mode, based on a first dictionary (D11). The second method performs second command voice recognition for second control commands relating to specific controls performed by the control processing unit (183) in call mode, based on a second dictionary (D12). The number of first control commands registered in the first dictionary (D11) is greater than the number of second control commands registered in the second dictionary (D12).

[0251] According to the above embodiment, the differentiation between the first and second methods is achieved by the difference in the number of control commands registered in the dictionary, thus enabling an improvement in the processing speed of speech recognition in a call state with a simpler configuration.

[0252] The information terminal (10) according to the fourth embodiment further comprises an answer operation unit (141) that receives an answer operation for answering a call to the information terminal (10) from another terminal in any one of the first to third embodiments. The voice recognition unit (182) switches to a state of voice recognition using the second method when it receives an answer operation.

[0253] According to the above configuration, switching to the second method is achieved through a response operation, thereby improving user convenience and enabling improvements to the voice recognition function.

[0254] The information terminal (10) according to the fifth embodiment further comprises at least one of an termination operation unit (142) and an unlocking operation unit (143) in any one of the first to fourth embodiments. The termination operation unit (142) receives a termination operation to end a call with another terminal. The unlocking operation unit (143) receives an unlocking operation to unlock an electric lock (201) provided on a door (front door 200) used for entering and exiting a facility (apartment building 5) where the information terminal (10) is installed. The voice recognition unit (182) switches to a state of voice recognition using the first method when it receives a termination operation or an unlocking operation while in a call with another terminal.

[0255] According to the above embodiment, switching to the first method is achieved by an termination operation or an unlocking operation, thereby improving user convenience and enabling improvements to the voice recognition function.

[0256] Regarding the information terminal (10) relating to the sixth embodiment, in any one of the first to fifth embodiments, at least one of an end command and an unlock command is set as a control command relating to the control performed by the control processing unit (183). The end command is a command for ending a call with another terminal. The unlock command is a command for unlocking an electric lock (201) installed on a door (front door 200) used for entering and exiting the facility (apartment building 5) where the information terminal (10) is installed. The voice recognition unit (182) switches to a state of voice recognition using the first method by voice recognition of the end command or the unlock command while in a call with another terminal (i.e., by successfully voice recognition of the end command or the unlock command).

[0257] According to the above embodiment, switching to the first method is achieved by uttering a termination command or an unlock command, thereby improving user convenience and enabling improvements to the voice recognition function.

[0258] With respect to the information terminal (10) according to the seventh aspect, in any one of the first to sixth aspects, a response command is set as a control command related to the control performed by the control processing unit (183). The response command is a command for answering a call to the information terminal (10) from another terminal. The voice recognition unit (182), while in a state of voice recognition using the first method, switches to a state of voice recognition using the second method by voice recognition of the response command (i.e., by successfully voice recognition of the response command).

[0259] According to the above configuration, switching to the second method is achieved by uttering a response command, thereby improving user convenience and enabling improvements to the speech recognition function.

[0260] The information terminal (10) according to the eighth embodiment further comprises, in the seventh embodiment, at least one of an termination operation unit (142) and an unlocking operation unit (143). The termination operation unit (142) receives a termination operation to end a call with another terminal. The unlocking operation unit (143) receives an unlocking operation to unlock an electric lock (201) provided on a door (front door 200) used for entering and exiting a facility (apartment building 5) where the information terminal (10) is installed. The voice recognition unit (182) switches to a state of voice recognition using the first method when it receives a termination operation or an unlocking operation while in a call with another terminal.

[0261] According to the above embodiment, switching to the first method is achieved by an termination operation or an unlocking operation, thereby improving user convenience and enabling improvements to the voice recognition function.

[0262] With respect to the information terminal (10) according to the ninth embodiment, in the seventh or eighth embodiment, at least one of an end command and an unlock command is set as a control command related to the control performed by the control processing unit (183). The end command is a command for ending a call with another terminal. The unlock command is a command for unlocking an electric lock (201) provided on a door (front door 200) used for entering and exiting the facility (apartment building 5) where the information terminal (10) is installed. The voice recognition unit (182) switches to a state of voice recognition using the first method by voice recognition of the end command or the unlock command while in a call with another terminal (i.e., by successfully voice recognition of the end command or the unlock command).

[0263] According to the above embodiment, switching to the first method is achieved by uttering a termination command or an unlock command, thereby improving user convenience and enabling improvements to the voice recognition function.

[0264] With respect to the information terminal (10) according to the tenth embodiment, in any one of the first to ninth embodiments, the voice recognition unit (182) performs voice recognition in the second method not only when in a call state with another terminal, but also when the information terminal (10) is being called by another terminal.

[0265] According to the above configuration, there is a high possibility that the processing speed related to speech recognition will improve not only in the call state with other terminals but also in the call state, thereby improving user convenience and enabling improvements to the speech recognition function.

[0266] With respect to the information terminal (10) according to the 11th embodiment, in the 10th embodiment, the voice recognition unit (182) switches to a state of voice recognition using the first method when the call is canceled.

[0267] According to the above configuration, since switching to the first method is achieved by canceling a call, user convenience is improved and improvements can be made to the voice recognition function.

[0268] With respect to the information terminal (10) according to the 12th embodiment, in any one of the 1st to 11th embodiments, the voice recognition unit (182) automatically switches to the state of voice recognition using the 1st method if a predetermined time elapses without voice input to the microphone (131) while in the state of voice recognition using the 2nd method.

[0269] According to the above configuration, switching to the first method can be achieved even without voice input, thereby improving user convenience and enabling improvements to the voice recognition function.

[0270] With respect to the information terminal (10) according to the 13th embodiment, in the second embodiment, the speech recognition unit (182) includes a first speech recognition unit (1821) that performs trigger speech recognition and a second speech recognition unit (1822) that performs command speech recognition. The second speech recognition unit (1822) is in an inactive state until the first speech recognition unit (1821) recognizes that a trigger word is included in the speech.

[0271] According to the above embodiment, the processing speed can be improved by separating the parts (processing circuits) that perform trigger voice recognition and command voice recognition, and by disabling the second voice recognition unit (1822) until it is recognized that a trigger word is included. In addition, the power consumption of the second voice recognition unit (1822) can be suppressed.

[0272] The intercom system (1) according to the 14th embodiment comprises an information terminal (10) according to any one of the 1st to 13th embodiments, and other terminals (lobby intercom 20, entrance unit 40).

[0273] According to the above embodiment, an intercom system (1) that can improve the voice recognition function can be provided.

[0274] An information terminal (10) according to the fifteenth aspect comprises a voice recognition unit (182), a control processing unit (183), a first communication unit (in the embodiment, the first communication unit 11 and the third communication unit 19), and a second communication unit (12). The voice recognition unit (182) performs voice recognition based on the voice input to a microphone (131). The first communication unit (in the embodiment, the first communication unit 11 and the third communication unit 19) performs communication including a call with other terminals (a lobby intercom 20 and an entrance slave unit 40). The second communication unit (12) communicates with a voice recognition system (A1) that performs command voice recognition on a control command related to control performed by the control processing unit (183). The information terminal (10) has a first mode and a second mode. In the first mode, the voice recognition unit (182) performs trigger voice recognition on whether or not a trigger word is included in the voice input to the microphone (131). Further, in the first mode, when the trigger word is included, communication is performed with the voice recognition system (A1) to cause the voice recognition system (A1) to perform command voice recognition. In the second mode, communication is performed with the voice recognition system (A1) to cause the voice recognition system (A1) to perform command voice recognition without causing the voice recognition unit (182) to perform trigger voice recognition. The information terminal (10) operates in the first mode in a standby state, and operates in the second mode in a call state with another terminal.

[0275] According to the above aspect, the information terminal (10) performs trigger voice recognition on its own device (locally), and causes the voice recognition system (A1) to perform command voice recognition. Therefore, the processing load on the information terminal (10) can be reduced. Further, since the information terminal (10) does not perform trigger voice recognition in a call state, the possibility that the processing speed related to voice recognition is improved increases, and user convenience can be improved. Furthermore, in a call state, the user does not need to utter a trigger word, which reduces the possibility that the trigger word uttered by the user is heard by the other party of the call. Therefore, a decrease in security can also be suppressed. As a result, the information terminal (10) has an advantage that the voice recognition function can be improved.

[0276] The voice recognition system (A1) according to the 16th embodiment comprises a communication unit (A11) that communicates with an information terminal (10) and a voice recognition unit (A12). When the voice recognition unit (A12) receives voice input to the microphone (131) from the information terminal (10) via the communication unit (A11), it performs voice recognition on the voice. The voice recognition unit (A12) has a first mode and a second mode. In the first mode, it performs trigger voice recognition to determine whether or not the voice input to the microphone (131) contains a trigger word. In the first mode, if a trigger word is contained, it performs command voice recognition for control commands related to the control performed by the control processing unit (183) of the information terminal (10). In the second mode, it performs command voice recognition without performing trigger voice recognition. The voice recognition system (A1) operates in a first mode when the information terminal (10) is in standby mode, and operates in a second mode when the information terminal (10) is in communication with other terminals (lobby intercom 20, door station 40).

[0277] According to the above configuration, since the speech recognition system (A1) performs both trigger speech recognition and command speech recognition, the processing load on the information terminal (10) can be reduced. Furthermore, since the speech recognition system (A1) does not perform trigger speech recognition during a call on the information terminal (10), the processing speed related to speech recognition is likely to improve, potentially enhancing user convenience. In addition, during a call, the user does not need to utter a trigger word, reducing the possibility that the trigger word uttered by the user may be heard by the other party on the call. Therefore, a decrease in security can also be mitigated. As a result, the speech recognition system (A1) has the advantage of being able to improve its speech recognition functionality.

[0278] A control method according to a 17th aspect is a control method for controlling an information terminal (10). The control method includes a voice recognition step, a control processing step, and a communication step. In the voice recognition step, voice recognition processing is performed based on a voice input to a microphone (131). In the control processing step, control is performed based on a voice recognition result obtained in the voice recognition step. In the communication step, communication including a call with another terminal (a lobby intercom 20, an entrance slave unit 40) is performed. The voice recognition processing performs voice recognition by a first method or a second method that is simpler than the voice recognition by the first method. The voice recognition processing performs voice recognition by the first method in a standby state, and performs voice recognition by the second method in a call state with another terminal.

[0279] According to the above aspect, it is possible to provide a control method capable of improving a voice recognition function.

[0280] A control method according to an 18th aspect is a control method for controlling an information terminal (10). The control method includes a voice recognition step, a control processing step, a first communication step, and a second communication step. In the voice recognition step, voice recognition processing is performed based on a voice input to a microphone (131). In the first communication step, communication including a call with another terminal (a lobby intercom 20, an entrance slave unit 40) is performed. In the second communication step, communication is performed with a voice recognition system that performs command voice recognition for a control command related to control performed in the control processing step. The information terminal (10) has a first mode and a second mode. In the first mode, trigger voice recognition is performed in the voice recognition step to determine whether or not a trigger word is included in a voice input to the microphone (131). In the first mode, when the trigger word is included, communication with the voice recognition system (A1) is performed in the second communication step to cause command voice recognition to be performed. In the second mode, communication with the voice recognition system (A1) is performed in the second communication step to cause command voice recognition to be performed, without performing trigger voice recognition in the voice recognition step. In the control method, the information terminal (10) is controlled to operate in the first mode in a standby state, and to operate in the second mode in a call state with another terminal.

[0281] According to the above embodiment, a control method can be provided that can improve the functionality of speech recognition.

[0282] The control method relating to the 19th aspect is a control method for controlling a speech recognition system (A1). The control method includes a communication step of communicating with an information terminal (10) and a speech recognition step. In the speech recognition step, when the system receives the voice input to the microphone (131) from the information terminal (10) in the communication step, it performs speech recognition processing on the voice. The speech recognition system (A1) has a first mode and a second mode. In the first mode, in the speech recognition step, trigger speech recognition is performed to determine whether or not the voice input to the microphone (131) contains a trigger word. In the first mode, if the trigger word is contained, command speech recognition is performed for control commands related to the control performed by the control processing unit (183) of the information terminal (10). In the second mode, in the speech recognition step, command speech recognition is performed without performing trigger speech recognition. In the control method, the information terminal (10) operates in a first mode when it is in standby mode, and operates in a second mode when the information terminal (10) is in communication with other terminals (lobby intercom 20, door station 40).

[0283] According to the above embodiment, a control method can be provided that can improve the functionality of speech recognition.

[0284] The program according to the 20th embodiment is a program that causes one or more processors to execute one of the control methods according to the 17th to 19th embodiments.

[0285] According to the above embodiment, it is possible to provide a function that can improve the functionality of speech recognition.

[0286] The configurations relating to the second to thirteenth aspects are not essential to the information terminal (10) relating to the first aspect and can be omitted as appropriate. [Explanation of symbols]

[0287] 1. Intercom System 10 Information terminals 182 Voice Recognition Unit 1821 First Speech Recognition Unit 1822 Second Speech Recognition Unit 183 Control Processing Unit 11. First Communications Department (Communications Department, First Communications Department) 12. Second Communications Department 131 Microphone 141 Response Operation Unit 142 Termination Operation Section 143 Unlocking operation unit 19. Third Communications Department (Communications Department, First Communications Department) 20. Lobby intercom (other terminals) 200 Entrance door (door) 201 Electric Lock 40. Door station (other terminal) 5. Apartment buildings (facilities) A1 Voice Recognition System A11 Communications Department A12 Voice Recognition Unit D11 First Dictionary D12 Second Dictionary

Claims

1. A voice recognition unit that performs voice recognition based on the sound input into the microphone, A control processing unit that performs control based on the speech recognition results of the speech recognition unit, A communications unit that handles communications, including calls with other terminals, Equipped with, The speech recognition unit performs speech recognition using the first method, or a second method that is simpler than the first method. The voice recognition unit performs voice recognition using the first method in standby mode and performs voice recognition using the second method in call mode with the other terminal. The first method is a method in which the control processing unit performs first command voice recognition for a first control command relating to a predetermined control performed in the standby state based on a first dictionary, The second method is a method in which the control processing unit performs second command speech recognition for a second control command relating to a specific control performed in the call state based on a second dictionary, The number of first control commands registered in the first dictionary is greater than the number of second control commands registered in the second dictionary. Information terminal.

2. The first method is, This method involves trigger speech recognition to determine whether the voice input to the microphone contains a trigger word, and if the trigger word is present, command speech recognition for the control command related to the control performed by the control processing unit. The second method described above is, This method performs command voice recognition without performing trigger voice recognition on the voice input to the microphone. The information terminal according to claim 1.

3. The system further comprises an answering operation unit that receives an answering operation for answering a call to the information terminal from another terminal, The voice recognition unit switches to a state of voice recognition using the second method upon receiving the response operation. The information terminal according to claim 1.

4. The system further comprises at least one of the following: an termination operation unit that receives an termination operation to end a call with the other terminal, and an unlocking operation unit that receives an unlocking operation to unlock an electric lock provided on a door for entering and exiting a facility where the information terminal is installed. The voice recognition unit switches to a state of voice recognition using the first method when it receives the termination operation or the unlock operation while in a call with the other terminal. The information terminal according to claim 1.

5. The control commands for the control processing unit include at least one of the following: an end command for ending a call with the other terminal and an unlock command for unlocking an electric lock provided on a door for entering and exiting the facility where the information terminal is installed. The voice recognition unit switches to a state of voice recognition using the first method when it recognizes the termination command or the unlock command by voice during a call with the other terminal. The information terminal according to claim 1.

6. As a control command relating to the control performed by the control processing unit, the information from the other terminal The terminal has a set response command for answering calls. The voice recognition unit, while in a state of voice recognition using the first method, switches to a state of voice recognition using the second method by recognizing the response command via voice. The information terminal according to claim 1.

7. The invention further comprises at least one of the following: an termination operation unit that receives an termination operation to terminate a call with the other terminal, and an unlocking operation unit that receives an unlocking operation to unlock an electric lock provided on a door for entering and exiting a facility where the information terminal is installed. The voice recognition unit switches to a state of voice recognition using the first method when it receives the termination operation or the unlock operation while in a call with the other terminal. The information terminal according to claim 6.

8. The control command for the control processing unit is set to include at least one of the following: an end command for ending a call with the other terminal and an unlock command for unlocking an electric lock provided on a door for entering and exiting the facility where the information terminal is installed. The voice recognition unit switches to a state of voice recognition using the first method when it recognizes the termination command or the unlock command by voice during a call with the other terminal. The information terminal according to claim 6.

9. The voice recognition unit performs voice recognition in the second manner not only when the other terminal is in a call state, but also when the other terminal is calling the information terminal. The information terminal according to claim 1.

10. The voice recognition unit switches to a state in which it performs voice recognition using the first method when the call is canceled. The information terminal according to claim 9.

11. The voice recognition unit, when in the state of voice recognition using the second method, automatically switches to the state of voice recognition using the first method if a predetermined time has elapsed without any voice input to the microphone. The information terminal according to claim 1.

12. The speech recognition unit comprises a first speech recognition unit that performs trigger speech recognition and a second speech recognition unit that performs command speech recognition, The second speech recognition unit remains in an inactive state until the first speech recognition unit recognizes that the trigger word is included in the standby state. The information terminal according to claim 2.

13. An information terminal according to any one of claims 1 to 12, The aforementioned other terminals, Equipped with, Intercom system.

14. A voice recognition unit that performs voice recognition based on the sound input to the microphone, Control processing unit and A first communication unit that handles communications, including calls with other terminals, A second communication unit communicates with a speech recognition system that performs speech recognition of control commands related to the control performed by the control processing unit, Equipped with, Whether or not the audio input to the microphone contains a trigger word A first mode in which the voice recognition unit performs trigger voice recognition, and if the trigger word is included, it communicates with the voice recognition system to perform command voice recognition, It has a second mode in which, instead of performing the trigger voice recognition by the voice recognition unit, it communicates with the voice recognition system to perform the command voice recognition, It operates in the first mode when in standby mode, and in the second mode when in call mode with the other terminal. Information terminal.

15. A communication unit that communicates with an information terminal, When the communication unit receives the audio input to the microphone from the information terminal, the voice recognition unit performs voice recognition on the audio, Equipped with, The aforementioned speech recognition unit, A first mode involves performing trigger voice recognition to determine whether the voice input to the microphone contains a trigger word, and if the trigger word is present, performing command voice recognition for control commands related to the control processing unit of the information terminal. A second mode in which command voice recognition is performed without trigger voice recognition, It has, The information terminal operates in the first mode when it is in standby mode, and operates in the second mode when it is in a call with another terminal. Voice recognition system.

16. A control method for controlling an information terminal, A speech recognition step that performs speech recognition processing based on the sound input to the microphone, A control processing step that performs control based on the speech recognition result of the speech recognition step, A communication step that includes making calls with other devices, Includes, The aforementioned speech recognition process performs speech recognition using the first method, or a second method that is simpler than the speech recognition of the first method. The aforementioned speech recognition process performs speech recognition using the first method in the standby state, and performs speech recognition using the second method in the state of a call with the other terminal. The first method is a method in which, in the control processing step, first command voice recognition for a first control command relating to a predetermined control performed in the standby state is performed based on a first dictionary, The second method is a method in which, in the control processing step, second command speech recognition for a second control command relating to a specific control performed in the call state is performed based on a second dictionary, The number of first control commands registered in the first dictionary is greater than the number of second control commands registered in the second dictionary. Control method.

17. A control method for controlling an information terminal, A speech recognition step that performs speech recognition processing based on the sound input to the microphone, Control processing step, A first communication step involves communication including making calls with other devices, A second communication step involves communicating with a speech recognition system that performs command speech recognition for control commands related to the control performed in the control processing step, Includes, The aforementioned information terminal is In the first mode, trigger speech recognition is performed in the speech recognition step to determine whether the voice input to the microphone contains a trigger word, and if the trigger word is found to be present, communication is performed in the second communication step to cause the speech recognition system to perform command speech recognition. A second mode is characterized in which, instead of performing the trigger voice recognition in the voice recognition step, communication with the voice recognition system is performed in the second communication step in order to perform the command voice recognition. It has, The information terminal is controlled to operate in the first mode when in standby mode and in the second mode when in call mode with the other terminal. Control method.

18. A control method for controlling a speech recognition system, Communication steps for communicating with an information terminal, When the audio input to the microphone is received from the information terminal in the communication step, the audio recognition step performs speech recognition processing on the audio, Includes, The aforementioned speech recognition system is In the aforementioned speech recognition step, trigger speech recognition is performed to determine whether or not the voice input to the microphone contains a trigger word, and if the trigger word is present, a first mode is performed to perform command speech recognition for a control command related to the control processing unit of the information terminal. In the aforementioned speech recognition step, there is a second mode in which the command speech recognition is performed without performing the trigger speech recognition, It has, The information terminal operates in the first mode when it is in standby mode, and operates in the second mode when it is in a call with another terminal. Control method.

19. A program for causing one or more processors to execute the control method described in any one of claims 16 to 18.

Images