JP6857012B2 - Alarm program and terminals using it - Google Patents

Description

The present invention relates to a technique for notifying a user of a disaster broadcast in a second language desired by a terminal user when the disaster broadcast is performed in the first language at the time of a disaster.

Conventionally, an emergency broadcasting facility is known as a facility for broadcasting the outbreak of a fire to the residents when a fire breaks out in a building. For example, Patent Document 1 describes an evacuation guidance device that broadcasts the details of a disaster, the place where the disaster occurs, and the evacuation method to the resident when a disaster such as a fire occurs in the building. A fire-related disaster broadcast is called an "emergency broadcast," which outputs a signal sound peculiar to an emergency broadcast and a predetermined voice message in a predetermined pattern, and depending on the situation, a detector alarm broadcast, a fire broadcast, or a non-fire. Output the broadcast. The signal sound is a signal that is continuously output in the order of the first sound and the second sound as one unit, and this is repeated three times in succession. The first sound is a single sound at 740 Hz for 0.5 seconds, and the second sound is a single sound at 494 Hz for 0.5 seconds (see FIG. 8A). The envelope has a waveform with a rising time of 0.1 seconds for the first sound and a waveform with a rising time of 0.4 seconds for the second sound. In the case of fire broadcasting, a sweep sound is further output. The sweep sound is a unit of signals that are continuous in the order of the third sound, the silence state for 0.5 seconds, the third sound, the silence state for 0.5 seconds, the third sound, and the silence state for 1.5 seconds. Is repeated 3 times in succession. The third sound is a sweep sound for 0.5 seconds from 300 Hz to 2 kHz, and the envelope is a square wave (see FIG. 8 (b)). The specifications of these signal sounds and sweep sounds are unique to Japan. Further, a disaster broadcast related to an earthquake is called an "emergency earthquake broadcast", and a signal sound peculiar to the emergency earthquake broadcast and a predetermined voice message are broadcast in a predetermined pattern (Non-Patent Document 1).

Japanese Unexamined Patent Publication No. 2-59999

Guidelines for emergency broadcasting equipment that responds to Earthquake Early Warnings, [0nline] Japan Electronics and Information Technology Industries Association, [Search on November 4, 2016], Internet <URL: http: //www.jeita.or. jp / japanese / standard / pdf / TTR-4701A_j.pdf>

However, conventional disaster broadcasting is based on Japan's own standards and is generally conducted only in Japanese, at most in several languages. Foreigners who could not understand these languages could not understand the contents. Moreover, even if disaster broadcasting is performed in a plurality of languages, it is not possible to hear the disaster by simultaneously broadcasting the plurality of languages, and it takes too much time to broadcast in a large number of languages in sequence. For this reason, the number of languages that can be used in disaster broadcasting is limited.

As a method of notifying foreigners who do not understand Japanese of the occurrence of a disaster, it is conceivable to notify by e-mail corresponding to multiple languages, but in that case, it is necessary to build a dedicated e-mail notification system on the facility side. This is costly and laborious. Also, on the user side, it takes time and effort to register or delete the e-mail address every time the facility is used. This is because if the e-mail address is not registered and deleted each time the facility is used, the disaster information e-mail of the facility will be received even outside the facility.

The present invention has been made in view of the above-mentioned problems, and when a disaster is broadcast in a first language at the time of a disaster, a preset first language is selected by a terminal user using computing technology. The purpose is to notify the user of the content of the disaster broadcast in two languages.

(1) In order to solve the above problems, the warning program according to the present invention is an acoustic input unit that collects the sound of a disaster broadcast performed in a first language by an emergency broadcasting facility and outputs an acoustic signal when a disaster occurs. And an alarm corresponding to the broadcast message of the disaster broadcast based on the acoustic pattern indicated by the acoustic signal output by the acoustic input unit of the terminal using the computing technology including the alarm output unit for outputting the alarm message. The alarm output includes an alarm message specifying unit that specifies an alarm message represented by a second language preset by the user of the terminal and an alarm message specified by the alarm message specifying unit. It is characterized in that it functions as an alarm unit to be output to the unit.

(2) In a preferred embodiment, in the alarm program according to the present invention, in the configuration described in (1), the disaster broadcast is detected based on the acoustic pattern indicated by the acoustic signal output by the acoustic input unit. It is characterized by being activated as an opportunity.

(3) In another preferred embodiment, the alarm program according to the present invention, in the configuration according to (1) or (2), has an acoustic signal indicating a message expressed in the first language or an acoustic signal indicated by the acoustic signal. Acquire the feature amount of, acquire the message represented by the first language by voice recognition based on the acoustic signal or the feature amount, translate the acquired message, and obtain the message represented by the second language. The alarm message identification unit sends an acoustic signal output by the acoustic input unit or an acoustic feature amount indicated by the acoustic signal to the voice translation device that acquires and returns, and the alarm message identification unit sends the alarm from the voice translation device. It is characterized by retrieving a message.

(4) The terminal according to the present invention has an acoustic input unit that collects the sound of a disaster broadcast performed in a first language by an emergency broadcasting facility and outputs an acoustic signal when a disaster occurs, and an alarm output that outputs an alarm message. A second alarm message corresponding to the broadcast message of the disaster broadcast, which is preset by the user of the own terminal, based on the sound pattern indicated by the unit and the sound signal output by the sound input unit. It is characterized by including an alarm message specifying unit that specifies an alarm message represented by a language, and an alarm unit that outputs an alarm message specified by the alarm message specifying unit to the alarm output unit.

According to any of the configurations (1) to (4), when a disaster broadcast is performed in the first language at the time of a disaster, the disaster broadcast is performed in a second language preset by the user of the terminal. Since the content can be notified to the user, the user can select a language to output a second language that he / she can understand so that he / she can understand the content of the disaster broadcast.

It is a block diagram which shows an example of the structure of terminal 1. It is a figure which shows an example of the alarm operation of the terminal 1, and shows the example of the operation which automatically starts the alarm operation. It is a flow chart which shows an example of the alarm operation of the terminal 1 and shows an example of the alarm operation after activation based on a signal sound. It is a block diagram which shows an example of the structure of terminal 1A. It is a flow chart which shows an example of the alarm operation of the terminal 1A. It is a block diagram which shows an example of the structure of the terminal 1B and the voice recognition server 2. It is a flow chart which shows an example of the alarm operation of the terminal 1B. It is a figure which shows the output example of a signal sound and a sweep sound in an emergency broadcast, (a) is a figure which shows a signal sound, (b) is a figure which shows the sweep sound respectively.

1. 1. First Embodiment 1-1. Configuration of Terminal 1 An example of the configuration of the terminal 1 according to the first embodiment will be described with reference to FIG. The terminal 1 is a terminal using computing technology carried by a user of a building, which is a fire prevention object, and specific examples thereof are a smartphone, a tablet terminal, and a wearable terminal. This terminal 1 collects the sound of the disaster broadcast in the first language performed by the emergency broadcasting equipment arranged in the building which is the fire prevention object, and the collected broadcast content is selected by the user. It has a function to notify in a preset second language. In the description of this embodiment, fire is assumed as the type of disaster, and emergency broadcasting is assumed as the type of disaster broadcasting. In addition, it is assumed that the emergency broadcast is performed in Japanese as the first language, and the notification to the user is performed in English as an example of the second language.

Here, the emergency broadcasting equipment is mainly equipment for notifying people in a building of the occurrence of a disaster and guiding evacuation in the event of a disaster, and in addition, various notifications associated therewith are given. The emergency broadcasting equipment automatically broadcasts disasters with voice alarm sounds in conjunction with the automatic fire alarm system. The emergency broadcasting equipment is composed of, for example, a starter, a speaker, an indicator light, an amplifier, an operation unit, a power supply, and wiring. Emergency broadcasting equipment is arranged in buildings such as hotels, inns, complex buildings, multi-tenant buildings, schools, libraries, underground malls, and quasi-underground malls.

Emergency broadcasting equipment has three basic operations. First, when the fire detector of the automatic fire alarm system is activated in response to a signal that the fire detector has been activated, the detector alarm broadcast is automatically performed. After that, if a signal from the transmitter or emergency telephone, a signal indicating that a fire detector other than the first fire detector has been activated, or if a predetermined time has passed, a fire is automatically fired. Broadcast. Fire broadcasting is also performed when the emergency start switch or fire broadcasting switch is pressed. On the other hand, when the non-fire broadcasting switch is pressed, non-fire broadcasting is performed.

Secondly, when it is activated by a transmitter or an emergency telephone, it automatically broadcasts a fire alarm immediately and automatically after the detector alert broadcast. However, in consideration of the use, scale, structure, and fire prevention management system of the fire prevention object, the sensor alarm broadcasting may be omitted and the fire broadcasting may be performed directly. If the fire broadcast is a false alarm and the non-fire broadcast switch is pressed, the non-fire broadcast is performed.

Thirdly, when the sensor alarm broadcasting is manually activated, if a predetermined time elapses, a fire broadcasting is automatically performed. Fire broadcasting is also performed when the emergency start switch or fire broadcasting switch is pressed. On the other hand, when the non-fire broadcasting switch is pressed, non-fire broadcasting is performed.

Emergency broadcasts related to fires performed by emergency broadcasting equipment are accompanied by signal sounds and sweep sounds (hereinafter collectively referred to as warning sounds), and there are three types: detector alert broadcasting, fire broadcasting, and non-fire broadcasting. There is a pattern. First, the detector alert broadcast is a broadcast that notifies a message that the fire detector has been activated, and is information that the fire detector has been activated, information that a fire has been confirmed, or related information. It is a broadcast that notifies the contents to be done. In addition, it may notify the location (floor) where the fire detector was activated. The structure of the broadcast is such that a continuous broadcast in the order of a warning sound called a signal sound (papopapopapo), a warning message, and a silent state for one second is set as one unit, and this is repeated twice or more in succession. Here, the warning message is, for example, the message "The fire detector (on the ○ floor) has been activated. The staff is checking it, so please be careful about the next broadcast (female voice)". ..

Next, the fire broadcast broadcasts a fire alarm message indicating that a fire has occurred, and is a broadcast that notifies information that a fire has occurred, evacuation guidance information, or related content. In addition, it may notify the location (floor) where the fire detector was activated. The broadcast consists of signal sound (papopapopapo), evacuation guidance message, 1 second silence, signal sound (papopapopapo), evacuation guidance message, 1 second silence, sweep sound (repeat fit, fit, fit 3 times). One unit is a continuous broadcast in the order of an alarm sound called, and this is continuously repeated for 10 minutes or more (for example, until fire extinguishing). Here, the evacuation guidance message is, for example, a message such as "It's a fire, it's a fire, a fire broke out (on the ○ floor). Please calm down and evacuate (male voice)".

Finally, the non-fire broadcast is a broadcast that notifies the above message that it was not a fire, and the information that the operation of the fire detector was a non-fire report or related content. It is a broadcast to notify. The structure of the broadcast is that a signal sound (papopapopapo), a correction message, and a continuous broadcast in the order of silence for 1 second are set as one unit, and this is repeated two or more times in succession. Here, the correction message is, for example, a message that "the operation of the fire detector was not abnormal as a result of confirmation. Please be assured (female voice)".

Return to the explanation of terminal 1. The terminal 1 includes a communication unit 11, a display unit 12, an operation unit 13, a storage unit 14, a control unit 15, a microphone 16, and a speaker 17.

[Communication Department]
The communication unit 11 is an interface for communicating with other devices via a communication line such as the Internet, a mobile communication network, or a wireless LAN. The communication unit 11 wirelessly connects to the communication line.

[Display]
The display unit 12 is a display device such as a liquid crystal display. The display unit 12 is an example of an alarm output unit that displays an alarm message under the control of the alarm unit 155, which will be described later. In the present embodiment, for convenience, various notification messages (for example, sensor alert message and non-fire message) will be included in the alarm message in addition to the fire alarm message. Further, for convenience, a program for outputting the fire alarm message and the various notification messages is referred to as an alarm program, and a means for outputting the fire alarm message and the various notification messages is referred to as an alarm output unit.

[Operation unit]
The operation unit 13 is an operation input device for a user carrying the terminal 1 to perform an input operation, and is, for example, a touch panel. The operation unit 13 detects the operation by the user and outputs the detected operation information to the control unit 15.

[Memory]
The storage unit 14 is a non-volatile storage device such as a flash memory or an SSD (Solid State Drive). The storage unit 14 stores the automatic start program 141. The automatic activation program 141 is a program for causing the terminal 1 using the computing technology to function as a signal sound recognition unit 151 and an activation unit 152, which will be described later.

Further, the storage unit 14 stores the alarm program 142. The alarm program 142 is a program for causing the terminal 1 using the computing technology to function as a sweep sound recognition unit 153, an alarm message identification unit 154, and an alarm unit 155, which will be described later. The alarm program 142 is activated by the activation unit 152, which will be described later, when a disaster broadcast is detected based on the acoustic pattern indicated by the acoustic signal output by the microphone 16. In the description of the present embodiment, an acoustic pattern of a warning sound (specifically, a signal sound) is assumed as a type of acoustic pattern, and an emergency broadcast related to a fire is assumed as a type of disaster broadcast. For example, when the terminal 1 is equipped with a camera, the alarm program 142 may be downloaded and installed from a server by reading a two-dimensional code posted on a bulletin board such as a poster attached to a building. Alternatively, it may be downloaded and installed from the application store. Further, the alarm program 142 may select a language desired by the user as a second language, select an alarm program 142 corresponding to the selected language, and download and install the alarm program 142. The alarm program 142 corresponding to the language may be downloaded and installed, and the language desired by the user may be selected and used as the second language. In this embodiment, English is selected as an example of the second language.

Further, the storage unit 14 stores a message database (“database” is referred to as “DB” in the drawings) 143. The message database 143 holds an alarm message in association with each selected language or a plurality of languages. Here, there are three types of alarm messages, a sensor alert message, a fire alarm message, and a non-fire message, and a voice message for voice output and a text message for display are prepared for each. .. Here, the sensor alert message is, for example, in English, the message "Attention please. The fire alarm is indicating a fire. We're now investigating the cause. Please wait for further information." The fire alarm message is, for example, in English, the message "There is a fire. Please evacuate as quickly as possible." A non-fire message is, for example, in English, the message "Attention please. A few minutes ago, we announced there may be a fire. However, this was a false alarm. Once again, there is no fire." The message database 143 may be formed at the same time when the alarm program 142 is installed.

In addition, the storage unit 14 uses a signal that is continuously output in the order of the first sound and the second sound, which is referred to for recognizing an emergency broadcast, as one unit, and repeats this three times in succession. Memorize the feature amount of. Here, the feature amount is a numerical value indicating the physical feature of the signal sound such as MFCC (mel frequency cepstrum coefficient) and LPC (linear prediction coefficient). In addition, the storage unit 14 stores the feature amount of the sweep sound, which is referred to for recognizing the fire broadcast. Here, the feature amount is a numerical value indicating the physical feature of the sweep sound such as MFCC and LPC. Further, when the alarm program 142 supports multiple languages, the storage unit 14 stores the setting of the second language selected as the language to be output at the time of execution. The second language setting stored here is referred to when specifying the language of the alarm message output to notify the user. The selection and setting of the second language may be performed, for example, at the time of downloading or installing the alarm program 142. In the description of this embodiment, it is assumed that English is set as the second language as the output language. Further, the storage unit 14 stores the activated flag as a variable for determining whether or not the alarm program 142 has been activated. This activated flag is initialized off, for example, at the end of alarm program 142. In addition, the storage unit 14 stores a count value indicating the number of times the emergency broadcast has been performed. This count value is initialized to "0" at the end of, for example, the alarm program 142. Further, the storage unit 14 stores the time measuring value of the time measuring means that measures the time from the initial value when the signal sound is recognized. When only a single second language is set as the language output by the alarm program 142, the storage unit 14 can omit the storage of the setting of the second language. For example, it is a case where the alarm program 142 corresponding to a single second language is installed by downloading or the like.

[Control unit]
The control unit 15 is a control device that executes a program stored in the storage unit 14 to control the operation of each unit of the terminal 1. The control unit 15 includes an arithmetic processing unit such as a CPU, and has a signal sound recognition unit 151, an activation unit 152, a sweep sound recognition unit 153, an alarm message identification unit 154, and an alarm unit 155 as functional configurations. Be prepared. These functions are realized by executing the automatic start program 141 and the alarm program 142.

The signal sound recognition unit 151 continuously acquires an acoustic signal from the microphone 16 when the automatic start program 141 stored in the storage unit 14 is executed. At this time, since there is an installation standard that the emergency broadcasting equipment must be arranged so as to have a predetermined sound pressure level (for example, the sound pressure level at a position 1 m from the floor is 75 dB or more), this installation standard is satisfied. The signal sound recognition unit 151 may acquire an acoustic signal having a volume equal to or higher than a predetermined volume to eliminate the noise sound. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal, and it is determined whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14 exceeds the threshold value. As a result of this determination, when the similarity exceeds the threshold value, it is recognized that the signal sound has been broadcast. On the other hand, if the similarity does not exceed the threshold value, the acoustic signal is acquired from the microphone 16 again. As for the method for extracting the feature amount, a well-known or known technique may be used.

When the signal sound is recognized by the signal sound recognition unit 151, the activation unit 152 refers to the activation flag stored in the storage unit 14 and determines whether or not the alarm program 142 has been activated. As a result of this determination, when the activated flag is set to off (that is, when the alarm program 142 has not been activated), the alarm program 142 stored in the storage unit 14 is activated. When the alarm program 142 is activated, the activated flag stored in the storage unit 14 is set to on, and the count value is incremented.

The sweep sound recognition unit 153 continuously acquires an acoustic signal from the microphone 16 when the alarm program 142 is activated by the activation unit 152. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal, and it is determined whether or not the similarity with the feature amount of the sweep sound stored in the storage unit 14 exceeds the threshold value. As a result of this determination, when the similarity exceeds the threshold value, it is recognized that the sweep sound has been broadcast. On the other hand, when the similarity does not exceed the threshold value, it is determined whether or not a predetermined time (for example, 50 seconds) has elapsed after the start of acquisition of the acoustic signal. As a result of this determination, it is recognized that the sweep sound has not been broadcast after a predetermined time has elapsed. On the other hand, if the predetermined time has not elapsed, the acoustic signal is acquired from the microphone 16 again. As for the method for extracting the feature amount, a well-known or known technique may be used. The predetermined time is set to the length of one unit or more of the fire broadcast.

The alarm message identification unit 154 is an alarm message corresponding to the broadcast message of the disaster broadcast based on the acoustic pattern indicated by the acoustic signal output by the microphone 16, and is a preset setting by the user of the terminal 1. Identify the alert message expressed in two languages. In the description of the present embodiment, an emergency broadcast related to a fire is assumed as the type of disaster broadcast, a warning sound (specifically, a sweep sound or a signal sound) pattern is assumed as the type of acoustic pattern, and English is used as the second language. Is assumed.

Specifically, when the sweep sound recognition unit 153 recognizes the sweep sound, the alarm message identification unit 154 acquires fire alarm messages for voice output and display from the message database 143. At that time, the fire alarm message corresponding to the set second language is acquired by referring to the setting of the second language stored in the storage unit 14. On the other hand, when a predetermined time elapses without the sweep sound being recognized by the sweep sound recognition unit 153, the most recent signal sound recognized by the signal sound recognition unit 151 is 1 with reference to the count value stored in the storage unit 14. Determine if it is the signal sound of the second emergency broadcast. As a result of this determination, if it is the signal sound of the first emergency broadcast, the sensor alert message for voice output and display is acquired from the message database 143. At that time, the sensor alert message corresponding to the set second language is acquired by referring to the setting of the second language stored in the storage unit 14. On the other hand, as a result of the determination, if it is not the signal sound of the first emergency broadcast, a non-fire message for voice output and display is acquired from the message database 143. At that time, the non-fire message corresponding to the set second language is acquired by referring to the setting of the second language stored in the storage unit 14.

The alarm unit 155 causes the alarm output unit to output the alarm message specified by the alarm message specifying unit 154. Specifically, the alarm message for voice output is output by voice from the speaker 17, and the warning message for display is displayed on the display unit 12.

[Microphone]
The microphone 16 is an acoustic input unit that collects sounds around the terminal 1. When a disaster occurs, the microphone 16 collects the sound of the disaster broadcast performed in the first language by the emergency broadcasting equipment and outputs an acoustic signal to the control unit 15. In the description of this embodiment, fire is assumed as the type of disaster, emergency broadcasting related to fire is assumed as the type of disaster broadcasting, and Japanese is assumed as the first language.

[Speaker]
The speaker 17 is an example of an alarm output unit that outputs an alarm sound or an alarm message by voice under the control of the alarm unit 155.

1-2. Operation of Terminal 1 Next, an example of alarm operation of terminal 1 will be described with reference to FIGS. 2 and 3.

First, an operation example of the automatic start program 141 will be described with reference to FIG. When the automatic start program 141 stored in the storage unit 14 is executed by the user's operation or automatically after the power of the terminal 1 is turned on, the alarm program 142 has already been started with reference to the started flag described later. Whether or not it is determined (Sa1). As a result of this determination, if the activated flag is set to on (that is, if the alarm program 142 has already been activated) (YES in Sa1), the system returns to the start time of the automatic activation program 141 and has been activated. This operation is repeated until the flag is turned off. On the other hand, as a result of this determination, if the activated flag is set to off (that is, if the alarm program 142 has not been activated) (NO in Sa1), the process proceeds to step Sa2. When the automatic start program 141 is executed for the first time, the started flag is set to off by the initial value, so the process proceeds to step Sa2, and the signal sound recognition unit 151 continuously acquires the acoustic signal from the microphone 16 ( Sa2). When the acoustic signal is acquired, the signal sound recognition unit 151 extracts the feature amount from the acquired acoustic signal (Sa3), and whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14 exceeds the threshold value. (Sa4). As a result of this determination, if the similarity does not exceed the threshold value (that is, if the signal sound is not recognized) (NO of Sa4), the signal sound recognition unit 151 returns to step Sa2, and the acoustic signal from the microphone 16 again. To get. On the other hand, as a result of this determination, when the similarity exceeds the threshold value (that is, when a signal sound is recognized) (YES of Sa4), the activation unit 152 is a counting value stored in the storage unit 14. Is incremented, and timekeeping by the timekeeping means is started (Sa5). Here, it is assumed that the count value is incremented for one emergency broadcast in which the signal sound is recognized (when the same broadcast is repeated a plurality of times, this is regarded as one emergency broadcast). Further, this timekeeping is initialized every time the signal sound of one emergency broadcast is recognized and timed, and the time after recognizing the signal sound is timed. Then, the activation unit 152 activates the alarm program 142 stored in the storage unit 14 (Sa6). After activating the alarm program 142, the auto-start program 141 returns to step Sa1. As will be described later, when the alarm program 142 is activated, the activated flag is set to on. Therefore, the automatic activation program 141 that activated the alarm program 142 repeats step Sa1 until the activated flag is turned off (Sa1). YES). As a result of the determination in step Sa1, when the activated flag is set to ON, since the alarm program 142 has already been activated, the automatic activation program 141 is stopped until the alarm program 142 is stopped. May be good.

Next, the operation of the activated alarm program 142 will be described with reference to FIG. When the alarm program 142 is activated, the activation unit 152 sets the activated flag stored in the storage unit 14 to on (Sa7), refers to the fire flag stored in the storage unit 14, and broadcasts a fire. It is determined whether or not it is recognized (Sa8). As a result of this determination, when the fire flag is set to ON (that is, when the alarm program 142 recognizes the fire broadcast) (YES of Sa8), the fire alarm described later is notified (Sa24). ). On the other hand, as a result of this determination, if the fire flag is set to off (that is, if the alarm program 142 does not recognize the fire broadcast) (NO of Sa8), the process proceeds to step Sa9. Since the fire flag when the alarm program 142 is activated is set to the initial value off, the process proceeds to step Sa9, and the sweep sound recognition unit 153 continuously acquires the acoustic signal from the microphone 16 (Sa9). When the acoustic signal is acquired, the sweep sound recognition unit 153 extracts the feature amount from the acquired acoustic signal (Sa10), and whether or not the similarity with the feature amount of the sweep sound stored in the storage unit 14 exceeds the threshold value. (Sa11). As a result of this determination, if the similarity exceeds the threshold value (that is, when a sweep sound is recognized) (YES of Sa11), the fire flag is set to ON (Sa12), and the process proceeds to step Sa13. On the other hand, as a result of the determination in step Sa11, if the similarity does not exceed the threshold value (that is, if the sweep sound is not recognized) (NO in Sa11), the process proceeds to step Sa13 without setting the fire flag, and the signal is signaled. The sound recognition unit 151 extracts a feature amount from the acquired acoustic signal (Sa13), and determines whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14 exceeds the threshold value (Sa14). .. As a result of this determination, if the similarity does not exceed the threshold value (that is, if the signal sound is not recognized) (NO of Sa14), the process proceeds to step Sa16. On the other hand, as a result of this determination, when the similarity exceeds the threshold value (that is, when a signal sound is recognized) (YES of Sa14), the count value stored in the storage unit 14 is incremented and timed. The value is initialized, the timing of the timing means is restarted from the initial value (Sa15), and the process proceeds to step Sa16. Then, each time the signal sound is recognized, it is initialized and timed, so that the time measured after recognizing the signal sound is referred to, and it is determined whether or not a predetermined time has elapsed (Sa16). This predetermined time is provided to determine that the most recently recognized signal sound is not due to a fire broadcast, and the sweep sound is output during the period in which the sweep sound should be output in the case of a fire broadcast. Judge by not having. Therefore, this predetermined time may be set to a time of at least one unit or more of repeated fire broadcasting. That is, in the case of a fire broadcast, after recognizing the signal sound, the evacuation guidance message, 1 second silence state, signal sound (papopapopapo), evacuation guidance message, 1 second silence state, sweep sound (fit, fit) , The fit is repeated 3 times), as long as it is longer than the total time of a series of continuous broadcasts. As a result of this determination, if the predetermined time has not elapsed (NO in Sa16), it cannot be determined that the most recently recognized signal sound is not a fire broadcast, so the process returns to step Sa8 and the sweep is performed by the time the predetermined time elapses. The determination (Sa11) of whether or not the sound is recognized is repeated. Then, when the sweep sound is recognized (YES in Sa11), the fire flag is turned on, so that it is determined that the fire was broadcast when returning to step Sa8 (YES in Sa8). Notify a fire alarm, which will be described later (Sa24). On the other hand, as a result of this determination, if the predetermined time has elapsed (YES in Sa16), the process proceeds to step Sa17. Here, if a sweep sound based on a fire broadcast is recognized immediately before the lapse of a predetermined time (YES in Sa11), a fire alarm must be notified, so whether or not the fire flag is turned on is determined by this. Judgment (Sa17). Then, when the fire flag is on (YES of Sa17), a fire alarm described later is notified (Sa24). On the other hand, as a result of this determination, when the fire flag is not turned on (NO of Sa17), the latest signal sound is not due to the fire broadcast, so by referring to the count value, the most recently recognized signal sound can be heard. It is determined whether the broadcast is due to a sensor alarm broadcast or a non-fire broadcast (Sa18). When the count value is 1 (YES in Sa18), since the most recently recognized signal sound is the first emergency broadcast, it is determined that it is due to the sensor alert broadcast, and the detector alert is notified. Then (Sa19), the process returns to step Sa8 in preparation for the next possible fire broadcast or non-fire broadcast. On the other hand, when the count value is not 1 (NO of Sa18), it is determined that the most recently recognized signal sound is due to the non-fire broadcast that was performed after the detector alert broadcast, and the non-fire report. Is notified (Sa20).

When notifying the sensor alert, the alarm message identification unit 154 acquires the sensor alert message for voice output and display from the message database 143. At that time, the alarm message identification unit 154 acquires an English sensor alert message by referring to the setting of the second language stored in the storage unit 14. When the sensor alarm message is acquired, the alarm unit 155 outputs the English sensor alarm message by voice from the speaker 17 and displays the English sensor alarm message on the display unit 12 (Sa19). Further, when notifying the non-fire report, the alarm message identification unit 154 acquires the non-fire message for voice output and display from the message database 143. At that time, the alarm message identification unit 154 acquires the English non-fire message by referring to the setting of the second language stored in the storage unit 14. When the non-fire message is acquired, the alarm unit 155 outputs the English non-fire message by voice from the speaker 17 and displays the English non-fire message on the display unit 12 (Sa20). Further, when notifying the fire alarm, the alarm message identification unit 154 acquires the fire alarm message for voice output and display from the message database 143. At that time, the alarm message identification unit 154 acquires an English fire alarm message by referring to the setting of the second language stored in the storage unit 14. When the fire alarm message is acquired, the alarm unit 155 outputs the English fire alarm message by voice from the speaker 17 and displays the English fire alarm message on the display unit 12 (Sa24).

The fire alarm message may be repeated until the alarm program 142 is stopped. Further, to stop the fire alarm, an alarm stop button (not shown) is displayed on the display unit 12 during the fire alarm, and the user of the terminal 1 operates the operation unit 13 corresponding to the alarm stop button (). YES of Sa25) The fire alarm may be stopped (Sa26). At this time, the audio output and display of the fire alarm message may be stopped, only the audio output of the fire alarm message may be stopped and the display may be left, or the output to be stopped may be appropriately selected. Further, the alarm program 142 is stopped by displaying an end button (not shown) on the display unit 12 during the fire alarm operation and operating the operation unit 13 corresponding to the end button by the user of the terminal 1 ( YES of Sa27), the alarm program 142 itself may be stopped (Sa28). Further, the non-fire message may at least retain its display until a new alarm message is output or the alarm program 142 is stopped. On the other hand, the voice output of the non-fire message may be stopped at one or more predetermined times, or can be stopped arbitrarily by the notification stop operation, and is repeated until the notification stop operation or the alarm program 142 is stopped. May be good. For example, an alarm stop button (not shown) is displayed on the display unit 12 displaying a non-fire message, and the user of the terminal 1 operates the operation unit 13 corresponding to the alarm stop button (YES of Sa21). ) The non-fire message may be stopped (Sa22). At this time, the voice output and display of the non-fire message may be stopped, only the voice output of the non-fire message may be stopped and the display may be left, or the output to be stopped may be appropriately selected. Further, to stop the alarm program 142, a display unit 12 displaying a non-fire message is displayed with an end button (not shown), and the user of the terminal 1 operates the operation unit 13 corresponding to the end button. (YES in Sa23), the alarm program 142 itself may be stopped (Sa28). The fire alarm may be stopped when a predetermined time (for example, 10 minutes) or more, which is assumed to have completed evacuation, has elapsed.
The above is the description of the alarm operation.

According to the terminal 1 described above, even a foreigner who cannot understand Japanese can understand the contents of the emergency broadcast by outputting an alarm message in a language that he / she can understand. Further, since the processing related to the above alarm operation is completed in the terminal 1, it is not necessary to construct a dedicated mail notification system on the facility side. In addition, there is no need for the user to register or delete the e-mail address. Moreover, even if the communication infrastructure does not function, it will not be affected.

1-3. Modification example of the first embodiment The above-mentioned first embodiment may be modified as follows.

1-3-1. Modification 1
After the alarm program 142 is activated, the alarm unit 155 may output the sensor alert message to the alarm output unit without waiting for the determination of whether or not the sweep sound is recognized (see Sa11). Then, when the signal sound is recognized again after the alarm program 142 is activated and the sweep sound is not recognized after the predetermined time has elapsed (YES in Sa16), the signal sound is heard. The non-fire message may be output to the alarm output unit without waiting for the determination of whether or not the broadcast is the first time (see Sa18). In the alarm operation according to this modified example, when the activated flag is turned on, the sensor alarm message is output to the alarm output unit, the signal sound is recognized again, and the sweep sound after a predetermined time has elapsed. If is not recognized, the non-fire message may be output to the alarm output unit, so that it is not necessary to store the count value indicating the number of times the emergency broadcast has been performed in the storage unit 14. If the first emergency broadcast is a fire broadcast instead of a detector alert broadcast, the detector alert message will be output once before the fire warning message is output, but if it is a fire broadcast Is not a problem because the sweep sound is recognized and a fire warning message is output (YES of Sa17).

2. Second Embodiment 2-1. Configuration of Terminal 1A An example of the configuration of terminal 1A according to the second embodiment will be described with reference to FIG. The terminal 1A according to the first embodiment is in the first embodiment in that it specifies an alarm message to be notified to the user based on the acoustic pattern of the broadcast message of the emergency broadcast instead of the acoustic pattern of the warning sound of the emergency broadcast. It is different from the terminal 1 according to the above.

The terminal 1A includes a communication unit 11A, a display unit 12A, an operation unit 13A, a storage unit 14A, a control unit 15A, a microphone 16A, and a speaker 17A.

[Communication Department]
The communication unit 11A is an interface for communicating with other devices via a communication line such as the Internet, a mobile communication network, or a wireless LAN. The communication unit 11A wirelessly connects to the communication line.

[Display]
The display unit 12A is a display device such as a liquid crystal display. The display unit 12A is an example of an alarm output unit that displays an alarm message under the control of the alarm unit 155A described later.

[Operation unit]
The operation unit 13A is an operation input device for a user carrying the terminal 1A to perform an input operation, and is, for example, a touch panel. The operation unit 13A detects an operation by the user and outputs the detected operation information to the control unit 15A.

[Memory]
The storage unit 14A is a non-volatile storage device such as a flash memory or SSD. The storage unit 14A stores the automatic start program 141A. The automatic activation program 141A is a program for causing the terminal 1A using the computing technology to function as a signal sound recognition unit 151A and an activation unit 152A, which will be described later.

Further, the storage unit 14A stores the alarm program 142A. The alarm program 142A is a program for causing the terminal 1A using the computing technology to function as a broadcast message identification unit 156, an alarm message identification unit 154A, and an alarm unit 155A, which will be described later. The alarm program 142A is activated by the activation unit 152A, which will be described later, when a disaster broadcast is detected based on the acoustic pattern indicated by the acoustic signal output by the microphone 16A. In the description of the present embodiment, an acoustic pattern of a warning sound (specifically, a signal sound) is assumed as the type of acoustic pattern, and emergency broadcasting is assumed as the type of disaster broadcasting. The acquisition method of the alarm program 142A may be the same as that of the alarm program 142 according to the first embodiment, for example.

Further, the storage unit 14A stores the message database 143A. The message database 143A holds each alarm message in association with the acoustic feature amount of the keyword unique to the broadcast message corresponding to the alarm message. Specifically, the sensor alert message is retained in association with the acoustic feature amount of "attention", which is a keyword unique to the sensor alert broadcast. In addition, a fire alarm message is retained in association with the acoustic feature amount of "fire", which is a keyword unique to fire broadcasting. In addition, the non-fire message is retained in association with the acoustic feature amount of "safety", which is a keyword unique to non-fire broadcasting. The keywords listed here are just examples, and other keywords may be adopted. Each alarm message is prepared for one selected language or a plurality of languages, and two types of messages, a voice message for voice output and a text message for display, are prepared. An example of each alarm message is as described in the description of the first embodiment.

In addition, the storage unit 14A stores the feature amount of the signal sound, which is referred to for recognizing the emergency broadcast. Here, the feature amount is a numerical value indicating the physical feature of the signal sound such as MFCC and LPC. Further, the storage unit 14A stores the setting of the second language as the language output when the alarm program 142A is executed. The setting of the second language stored here is referred to when specifying the language of the alarm message notified to the user. The setting of the second language may be performed, for example, when the alarm program 142A is installed. In the description of this embodiment, it is assumed that English is set as the second language. Further, the storage unit 14A stores the activated flag as a variable for determining whether or not the alarm program 142A is activated. This activated flag is initialized off, for example, at the end of alarm program 142A.

[Control unit]
The control unit 15A is a control device that executes a program stored in the storage unit 14A to control the operation of each unit of the terminal 1A. The control unit 15A includes an arithmetic processing unit such as a CPU, and has a signal sound recognition unit 151A, an activation unit 152A, a broadcast message identification unit 156, an alarm message identification unit 154A, and an alarm unit 155A as functional configurations. Be prepared. Among these functions, the signal sound recognition unit 151A and the activation unit 152A are realized by executing the automatic activation program 141A, and the broadcast message identification unit 156, the alarm message identification unit 154A, and the alarm unit 155A provide the alarm program 142A. It is realized by executing.

The signal sound recognition unit 151A continuously acquires an acoustic signal from the microphone 16A when the automatic start program 141A stored in the storage unit 14A is executed. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal, and it is determined whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14A exceeds the threshold value. As a result of this determination, when the similarity exceeds the threshold value, it is recognized that the signal sound has been broadcast. On the other hand, if the similarity does not exceed the threshold value, the acoustic signal is acquired from the microphone 16A again. As in the first embodiment, the signal sound recognition unit 151A may acquire an acoustic signal having a predetermined volume or higher to eliminate the noise sound. Further, as a method for extracting the feature amount, a well-known or known technique may be used.

When the signal sound is recognized by the signal sound recognition unit 151A, the activation unit 152A determines whether or not the alarm program 142A has been activated by referring to the activation flag stored in the storage unit 14A. As a result of this determination, when the activated flag is set to off (that is, when the alarm program 142A has not been activated), the alarm program 142A stored in the storage unit 14A is activated. When the alarm program 142A is activated, the activated flag stored in the storage unit 14A is set to on.

When the alarm program 142A is activated by the activation unit 152A, the broadcast message identification unit 156 acquires the acoustic signal of the broadcast message to be broadcast after the signal sound from the microphone 16. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal and compared with the acoustic feature amount of each keyword stored in the storage unit 14A to determine which keyword is included in the broadcast message. As a result of this determination, if the keyword "caution" is included, it is recognized that the detector alert broadcast has been performed. If the keyword "fire" is included, it is recognized that a fire has been broadcast. If the keyword "safety" is included, it is recognized that a non-fire broadcast has been performed. As for the method for extracting the feature amount and the method for detecting the keyword based on the acoustic feature amount, a well-known technique may be used.

The alarm message identification unit 154A is an alarm message corresponding to the broadcast message of the disaster broadcast based on the acoustic pattern indicated by the acoustic signal output by the microphone 16A, and is a preset setting by the user of the terminal 1A. Identify the alert message expressed in two languages. In the description of the present embodiment, emergency broadcasting is assumed as the type of disaster broadcasting, the acoustic pattern of the broadcast message is assumed as the type of acoustic pattern, and English is assumed as the second language.

Specifically, when the broadcast message identification unit 156 recognizes the sensor alert broadcast, the alarm message identification unit 154A is associated with the acoustic feature amount of the keyword "attention" from the message database 143A. Get detector alert messages for output and display. At that time, the sensor alert message corresponding to the set second language is acquired by referring to the setting of the second language stored in the storage unit 14A. Alternatively, when the broadcast message identification unit 156 recognizes the fire broadcast, the fire alarm message for voice output and display, which is associated with the acoustic feature amount of the keyword "fire", is acquired from the message database 143A. At that time, the fire alarm message corresponding to the set second language is acquired by referring to the setting of the second language stored in the storage unit 14A. Alternatively, when the broadcast message identification unit 156 recognizes the non-fire broadcast, the non-fire message for audio output and display, which is associated with the acoustic feature amount of the keyword "safety", is acquired from the message database 143A. .. At that time, the non-fire message corresponding to the set second language is acquired by referring to the setting of the second language stored in the storage unit 14A.

The alarm unit 155A causes the alarm output unit to output the alarm message specified by the alarm message specifying unit 154A. Specifically, the alarm message for voice output is output by voice from the speaker 17A, and the warning message for display is displayed on the display unit 12A.

[Microphone]
The microphone 16A is an acoustic input unit that collects sounds around the terminal 1A. When a disaster occurs, the microphone 16A collects the sound of the disaster broadcast performed in the first language by the emergency broadcasting equipment and outputs an acoustic signal to the control unit 15A. In the description of this embodiment, fire is assumed as the type of disaster, emergency broadcasting is assumed as the type of disaster broadcasting, and Japanese is assumed as the first language.

[Speaker]
The speaker 17A is an example of an alarm output unit that outputs an alarm sound or an alarm message by voice under the control of the alarm unit 155A.

2-2. Operation of Terminal 1A Next, an example of alarm operation of terminal 1A will be described with reference to FIG.

After the power of the terminal 1A is turned on, when the automatic start program 141A stored in the storage unit 14A is executed by the user's operation or automatically, the signal sound recognition unit 151A continuously outputs an acoustic signal from the microphone 16A. Acquire (Sb1). When the acoustic signal is acquired, the signal sound recognition unit 151A extracts the feature amount from the acquired acoustic signal (Sb2), and whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14A exceeds the threshold value. (Sb3). As a result of this determination, if the similarity does not exceed the threshold value (that is, if the signal sound is not recognized) (NO in Sb3), the signal sound recognition unit 151A returns to step Sb1 and returns to the acoustic signal from the microphone 16A. To get. On the other hand, as a result of this determination, when the similarity exceeds the threshold value (that is, when a signal sound is recognized) (YES of Sb3), the activation unit 152A is stored in the storage unit 14A and has been activated. With reference to the flag, it is determined whether or not the alarm program 142A has been activated (Sb4). As a result of this determination, if the activated flag is set to on (that is, if the alarm program 142A has been activated) (YES in Sb4), steps Sb5 and Sb6 are omitted, and step Sb7 is executed. Will be done. On the other hand, as a result of this determination, when the activated flag is set to off (that is, when the alarm program 142A has not been activated) (NO of Sb4), the activated unit 152A stores in the storage unit 14A. The alarm program 142A is activated (Sb5). After activating the alarm program 142A, the activation unit 152A sets the activated flag to ON (Sb6).

When the alarm program 142A is activated, the broadcast message identification unit 156 acquires the acoustic signal of the broadcast message to be broadcast after the signal sound from the microphone 16A (Sb7). When the acoustic signal is acquired, the broadcast message identification unit 156 extracts the feature amount from the acquired acoustic signal (Sb8) and compares it with the acoustic feature amount of each keyword stored in the storage unit 14A. It is determined whether or not the keyword of is included (Sb9). As a result of this determination, if the keyword "caution" is included (that is, when the sensor alert broadcast is recognized) (Sb9 "sensor alert broadcast"), the alarm message identification unit 154A Acquires the sensor alert message for voice output and display, which is associated with the acoustic feature amount of the keyword "attention", from the message database 143A. At that time, the alarm message identification unit 154A acquires an English sensor alert message by referring to the setting of the second language stored in the storage unit 14A. When the sensor alarm message is acquired, the alarm unit 155A outputs the English sensor alarm message by voice from the speaker 17A and displays the English sensor alarm message on the display unit 12A (Sb10). After the detector alert message is output, step Sb1 is executed again.

As a result of the determination in step Sb9, if the keyword "fire" is included (that is, when a fire broadcast is recognized) ("fire broadcast" in Sb9), the alarm message identification unit 154A uses the message database. From 143A, a fire alarm message for audio output and display, which is associated with the acoustic feature amount of the keyword "fire", is acquired. At that time, the alarm message identification unit 154A acquires an English fire alarm message by referring to the setting of the second language stored in the storage unit 14A. When the fire alarm message is acquired, the alarm unit 155A outputs the English fire alarm message by voice from the speaker 17A and displays the English fire alarm message on the display unit 12A (Sb11). After the fire alarm message is output, the alarm program 142A may be repeated until it is stopped. To stop the fire alarm, a display unit 12A during the fire alarm is displayed with an alarm stop button (not shown), and the user of the terminal 1A operates the operation unit 13A corresponding to the alarm stop button to cause a fire. The alarm may be stopped. At this time, the audio output and display of the fire alarm message may be stopped, only the audio output of the fire alarm message may be stopped and the display may be left, or the output to be stopped may be appropriately selected. Further, to stop the alarm program 142A, an end button (not shown) is displayed on the display unit 12A during the fire alarm operation, and the user of the terminal 1A operates the operation unit 13A corresponding to the end button to give an alarm. The program 142A itself may be stopped.

As a result of the determination in step Sb9, if the keyword "safety" is included (that is, when a non-fire broadcast is recognized) ("non-fire broadcast" in Sb9), the alarm message identification unit 154A From the message database 143A, non-fire messages for voice output and display, which are associated with the acoustic features of the keyword "safety", are acquired. At that time, the alarm message identification unit 154A acquires the English non-fire message by referring to the setting of the second language stored in the storage unit 14A. When the non-fire message is acquired, the alarm unit 155A outputs the English non-fire message by voice from the speaker 17A and displays the English non-fire message on the display unit 12A (Sb12). After the output of the non-fire message, the output operation may be terminated, or the alarm program 142A may be repeated until it is stopped. Further, the non-fire message may at least retain its display until a new alarm message is output or the alarm program 142A is stopped. On the other hand, the voice output of the non-fire message may be stopped at one or more predetermined times, or can be stopped arbitrarily by the notification stop operation, and is repeated until the notification stop operation or the alarm program 142A is stopped. May be good. For example, a non-fire message is displayed by displaying an alarm stop button (not shown) on the display unit 12A displaying the non-fire message, and the user of the terminal 1A operates the operation unit 13A corresponding to the alarm stop button to display the non-fire message. You may want to stop it. At this time, the voice output and display of the non-fire message may be stopped, only the voice output of the non-fire message may be stopped and the display may be left, or the output to be stopped may be appropriately selected. Further, to stop the alarm program 142A, a display unit 12A displaying a non-fire message is displayed with an end button (not shown), and the user of the terminal 1A operates the operation unit 13A corresponding to the end button. The alarm program 142A itself may be stopped by. As described above, the operation related to the alarm stop in the second embodiment is the same as that in the first embodiment.
The above is the description of the alarm operation.

According to the terminal 1A described above, as in the terminal 1 according to the first embodiment, even a foreigner who cannot understand Japanese can understand the contents of the emergency broadcast. In addition, since the processing related to the above alarm operation is completed in the terminal 1A, it is not necessary to build a dedicated e-mail notification system on the facility side, and the user side registers or deletes the e-mail address. It does not take time and effort. Moreover, even if the communication infrastructure does not function, it will not be affected.

2-3. Modification example of the second embodiment The above-mentioned second embodiment may be modified as follows. The following modification may be combined with one or more other modification.

2-3-1. Modification 1
The above-mentioned broadcast message identification unit 156 compares the feature amount of the broadcast message broadcast after the signal sound with the acoustic feature amount of each broadcast message as a whole instead of the acoustic feature amount of each keyword, so that the emergency broadcast can be performed. The type may be determined. In that case, the message database 143A holds each alarm message in association with the acoustic feature amount of the entire broadcast message corresponding to the alarm message. The broadcast message identification unit 156 determines which of the acoustic features of each broadcast message stored in the message database 143A is the most similar to the feature amount of the broadcast message broadcast after the signal sound. The alarm message identification unit 154A acquires an alarm message associated with an acoustic feature amount most similar to the feature amount of the broadcast message by the broadcast message identification unit 156.

2-3-2. Modification 2
All or part of the processing by the broadcast message identification unit 156 and the alarm message identification unit 154A may be performed by an external server connected to the terminal 1A via a communication line. For example, the broadcast message identification unit 156 transmits the feature amount of the broadcast message to be broadcast after the signal sound to the external server, asks the external server to determine the keyword included in the broadcast message, and receives the keyword. You may do so. At that time, the broadcast message identification unit 156 may transmit the acoustic signal of the broadcast message to the external server and have the external server extract the feature amount. As another example, the broadcast message identification unit 156 transmits the feature amount of the broadcast message to be broadcast after the signal sound to the external server, determines the keyword included in the broadcast message on the external server side, and corresponds to the keyword. The alarm message may be specified so that the alarm message identification unit 154A receives the specified alarm message or its identifier. At that time, the broadcast message identification unit 156 may transmit the acoustic signal of the broadcast message to the external server and have the external server extract the feature amount.

2-3-3. Modification 3
The terminal 1A may include the sweep sound recognition unit 153 according to the first embodiment, and may refer to the result of the determination by the sweep sound recognition unit 153 for the identification of the fire broadcast.

3. 3. Third Embodiment 3-1. Configuration of Terminal 1B An example of the configuration of terminal 1B according to the third embodiment will be described with reference to FIG. The terminal 1B according to the first embodiment relates to the first embodiment in that the terminal 1B according to the first embodiment specifies a message to be notified to the user based on the content of the broadcast message of the emergency broadcast instead of the acoustic pattern of the warning sound of the emergency broadcast. It is different from terminal 1. Here, the message to be notified to the user is specified by the voice recognition server 2 voice recognition and translation of the broadcast message.

The terminal 1B includes a communication unit 11B, a display unit 12B, an operation unit 13B, a storage unit 14B, a control unit 15B, a microphone 16B, and a speaker 17B.

[Communication Department]
The communication unit 11B is an interface for communicating with the voice recognition server 2 via a communication line L such as the Internet, a mobile communication network, or a wireless LAN. The communication unit 11A wirelessly connects to the communication line L.

[Display]
The display unit 12B is a display device such as a liquid crystal display. The display unit 12B is an example of an alarm output unit that displays an alarm message under the control of the alarm unit 155B described later.

[Operation unit]
The operation unit 13B is an operation input device for a user carrying the terminal 1B to perform an input operation, and is, for example, a touch panel. The terminal 1B detects the operation by the user and outputs the detected operation information to the control unit 15B.

[Memory]
The storage unit 14B is a non-volatile storage device such as a flash memory or SSD. The storage unit 14B stores the automatic start program 141B. The automatic start program 141B is a program for causing the terminal 1B, which is a computer, to function as a signal sound recognition unit 151B and a start unit 152B, which will be described later.

Further, the storage unit 14B stores the alarm program 142B. The alarm program 142B is a program for causing the terminal 1B, which is a computer, to function as a message identification unit 154B, an alarm unit 155B, and a sweep sound stop recognition unit 157, which will be described later. The alarm program 142B is activated by the activation unit 152B, which will be described later, when a disaster broadcast is detected based on the acoustic pattern indicated by the acoustic signal output by the microphone 16B. In the description of the present embodiment, an acoustic pattern of a warning sound (specifically, a signal sound) is assumed as the type of acoustic pattern, and emergency broadcasting is assumed as the type of disaster broadcasting. The acquisition method of the alarm program 142B may be the same as that of the alarm program 142 according to the first embodiment, for example.

In addition, the storage unit 14B stores the feature amount of the signal sound referred to for recognizing the emergency broadcast. Here, the feature amount is a numerical value indicating the physical feature of the signal sound such as MFCC and LPC. Further, the storage unit 14B stores the setting of the second language as the language output when the alarm program 142B is executed. The setting of the second language stored here is referred to when specifying the language of the alarm message notified to the user. The second language may be set, for example, when the alarm program 142B is installed. In the description of this embodiment, it is assumed that English is set as the second language. Further, the storage unit 14B stores the activated flag as a variable for determining whether or not the alarm program 142B is activated. This activated flag is initialized off, for example, at the end of alarm program 142B.

[Control unit]
The control unit 15B is a control device that executes a program stored in the storage unit 14B to control the operation of each unit of the terminal 1B. The control unit 15B includes an arithmetic processing unit such as a CPU, and has a signal sound recognition unit 151B, an activation unit 152B, a message identification unit 154B, an alarm unit 155B, and a sweep sound stop recognition unit 157 as functional configurations. Be prepared. Among these functions, the signal sound recognition unit 151B and the activation unit 152B are realized by executing the automatic activation program 141B, and the message identification unit 154B, the alarm unit 155B, and the sweep sound stop recognition unit 157 provide the alarm program 142B. It is realized by executing.

The signal sound recognition unit 151B continuously acquires an acoustic signal from the microphone 16B when the automatic start program 141B stored in the storage unit 14B is executed. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal, and it is determined whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14B exceeds the threshold value. As a result of this determination, when the similarity exceeds the threshold value, it is recognized that the signal sound has been broadcast. On the other hand, if the similarity does not exceed the threshold value, the acoustic signal is acquired from the microphone 16B again. As for the method for extracting the feature amount, a well-known or known technique may be used.

When the signal sound is recognized by the signal sound recognition unit 151B, the activation unit 152B refers to the activation flag stored in the storage unit 14B and determines whether or not the alarm program 142B has been activated. As a result of this determination, when the activated flag is set to off (that is, when the alarm program 142B has not been activated), the alarm program 142B stored in the storage unit 14B is activated. When the alarm program 142B is activated, the activated flag stored in the storage unit 14B is set to on.

The message identification unit 154B corresponds to the alarm message identification unit 154 and 154A described above, and is based on the content of the broadcast message (an example of the acoustic pattern according to the present invention) indicated by the acoustic signal output by the microphone 16B. The translated message represented by the second language set in advance by the user of the terminal 1B is specified. This translated message corresponds to the alarm message described above. In the description of the present embodiment, an emergency broadcast related to a fire is assumed as a type of disaster broadcast, and in addition to a predetermined message automatically output from the emergency broadcast equipment as a broadcast message, a message broadcast by the operator using a microphone is displayed. Assuming, English is assumed as the second language.

Specifically, when the alarm program 142B is activated by the activation unit 152B, the message identification unit 154B acquires the acoustic signal of the broadcast message to be broadcast after the signal sound from the microphone 16B. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal, and the voice translation request including the extracted feature amount data is transmitted to the voice recognition server 2 by using the communication unit 11B. At that time, the data indicating the second language stored in the storage unit 14B is included in the speech translation request as the data indicating the language of the translation destination. When an English translation message for voice output and display is transmitted from the voice recognition server 2 as a reply to the voice translation request, the translation message is received via the communication unit 11B. As for the method for extracting the feature amount, a well-known or known technique may be used.

Further, when the translated message is output by the alarm output unit, the message specifying unit 154B determines whether or not the sensor alert message is output. This determination is performed, for example, by determining whether or not the output alarm message contains a keyword (for example, "investigating") unique to the sensor alert message. As a result of this determination, if the sensor alarm message is not output, it is further determined whether or not the fire alarm message is output. This determination is performed, for example, by determining whether or not the output translated message contains a keyword (for example, "evacuate") unique to the fire warning message.

Further, when the message identification unit 154B determines that the sweep sound stop recognition unit 157 has stopped broadcasting the sweep sound for a predetermined time or longer (in other words, when the fire broadcast is stopped and the operator's microphone broadcasts). ), The acoustic signal of the broadcast message to be broadcast after the broadcast of the sweep sound is stopped is acquired from the microphone 16B. When the acoustic signal is acquired, the feature amount is extracted from the acquired acoustic signal, and the voice translation request including the extracted feature amount data is transmitted to the voice recognition server 2 by using the communication unit 11B. At that time, the data indicating the second language stored in the storage unit 14B is included in the speech translation request as the data indicating the language of the translation destination. When an English translation message for voice output and display is transmitted from the voice recognition server 2 as a reply to the voice translation request, the translation message is received via the communication unit 11B.

The alarm unit 155B causes the alarm output unit to output the translated message specified by the message identification unit 154B. Specifically, the alarm message for voice output is output by voice from the speaker 17B, and the warning message for display is displayed on the display unit 12B.

When the alarm output unit outputs a fire alarm message, the sweep sound stop recognition unit 157 indicates whether or not the sweep sound broadcast has been stopped for a predetermined time or longer (in other words, the fire broadcast is stopped and the operator's microphone broadcasts. Whether or not was performed) is determined. For example, the sweep sound stop recognition unit 157 continuously acquires an acoustic signal from the microphone 16B, extracts a feature amount, and compares it with the feature amount of the sweep sound stored in the storage unit 14B to broadcast the sweep sound. Judge the stop of. As a result of this determination, if the broadcast of the sweep sound has not been stopped for a predetermined time or longer, it is further determined whether or not a predetermined time (for example, 30 seconds) has elapsed after the activation of the alarm program 142B.

[Microphone]
The microphone 16B is an acoustic input unit that collects sounds around the terminal 1B. When a disaster occurs, the microphone 16B collects the sound of the disaster broadcast performed in the first language by the emergency broadcasting equipment and outputs an acoustic signal to the control unit 15B. In the description of this embodiment, fire is assumed as the type of disaster, emergency broadcasting is assumed as the type of disaster broadcasting, and Japanese is assumed as the first language.

[Speaker]
The speaker 17B is an example of an alarm output unit that outputs an alarm sound or a translated message by voice under the control of the alarm unit 155B.

3-2. Configuration of Voice Recognition Server 2 An example of the configuration of the voice recognition server 2 according to the third embodiment will be described with reference to FIG. The voice recognition server 2 acquires the acoustic features indicated by the acoustic signal indicating the message represented by the first language from the terminal 1B, and based on the acquired features, voice recognizes the message represented by the first language. It is a voice translation device that acquires, translates the acquired message, acquires a message expressed in a second language, and returns it to the terminal 1B. In the description of this embodiment, Japanese is assumed as the first language and English is assumed as the second language. The voice recognition server 2 includes a communication unit 21, a storage unit 22, and a control unit 23.

[Communication Department]
The communication unit 21 is an interface for communicating with the terminal 1B via the communication line L. The communication unit 21 connects to the communication line L wirelessly or by wire.

[Memory]
The storage unit 22 is a non-volatile storage device such as a ROM or a hard disk. The storage unit 22 stores a speech translation program executed by the control unit 23. Further, the storage unit 22 stores an acoustic model such as an HMM (Hidden Markov Model), a language model such as an N-gram model, and a word dictionary used for speech recognition processing. Further, the storage unit 22 stores a translation model such as a logarithmic linear model used in the machine translation process.

[Control unit]
The control unit 23 is a control device that executes a program stored in the storage unit 22 to control the operation of each unit of the voice recognition server 2. The control unit 23 includes an arithmetic processing unit such as a CPU, and has a functional configuration of a voice translation request reception unit 231, a voice recognition unit 232, a machine translation unit 233, a voice synthesis unit 234, and a message transmission unit 235. To be equipped.

The voice translation request reception unit 231 receives the voice translation request transmitted from the terminal 1B via the communication unit 21.

When the voice recognition unit 232 acquires the data of the feature amount included in the voice translation request received by the voice translation request reception unit 231, the voice recognition unit 232 uses the acoustic model, the language model, and the word dictionary stored in the storage unit 22 to perform voice recognition. To identify the character string that represents the broadcast message. As for the voice recognition method, a well-known or known technique may be used.

The machine translation unit 233 machine-translates the character string representing the broadcast message specified by the voice recognition unit 232 into the language specified by the voice translation request received by the voice translation request reception unit 231 and translates the message for display. Generate (text data). In the description of this embodiment, it is assumed that an English translation message is generated. As for the machine translation method, a well-known or known technique may be used.

When the voice synthesis unit 234 acquires the translated message generated by the machine translation unit 233, the voice synthesis unit 234 performs voice synthesis using the acoustic model stored in the storage unit 22 and generates a translated message (voice data) for voice output. .. As for the method of speech synthesis, a well-known technique may be used.

The message transmission unit 235 returns the translation message for display generated by the machine translation unit 233 and the translation message for voice output generated by the voice synthesis unit 234 to the terminal 1B via the communication unit 21.

3-3. Operation of Terminal 1B Next, an example of alarm operation of terminal 1B will be described with reference to FIG. 7.

After the power of the terminal 1B is turned on, when the automatic start program 141B stored in the storage unit 14B is executed by the user's operation or automatically, the signal sound recognition unit 151B continuously outputs an acoustic signal from the microphone 16B. Acquire (Sc1). When the acoustic signal is acquired, the signal sound recognition unit 151B extracts the feature amount from the acquired acoustic signal (Sc2), and whether or not the similarity with the feature amount of the signal sound stored in the storage unit 14B exceeds the threshold value. (Sc3). As a result of this determination, if the similarity does not exceed the threshold value (that is, if the signal sound is not recognized) (NO of Sc3), the signal sound recognition unit 151B returns to step Sc1 and the acoustic signal is again emitted from the microphone 16B. To get. On the other hand, as a result of this determination, when the similarity exceeds the threshold value (that is, when a signal sound is recognized) (YES of Sc3), the activation unit 152B is stored in the storage unit 14B. With reference to the flag, it is determined whether or not the alarm program 142B has been activated (Sc4). As a result of this determination, if the activated flag is set to on (that is, if the alarm program 142B has already been activated) (YES in Sc4), steps Sc5 and Sc6 are omitted and step Sc7 is executed. Will be done. On the other hand, as a result of this determination, when the activated flag is set to off (that is, when the alarm program 142B has not been activated) (NO of Sc4), the activated unit 152B stores in the storage unit 14B. The alarm program 142B is activated (Sc5). After activating the alarm program 142B, the activation unit 152B sets the activated flag to ON (Sc6).

When the alarm program 142B is activated, the message identification unit 154B acquires the acoustic signal of the broadcast message to be broadcast after the signal sound from the microphone 16B (Sc7). When the acoustic signal is acquired, the message identification unit 154B extracts a feature amount from the acquired acoustic signal (Sc8), and sends a voice translation request including the extracted feature amount data to the voice recognition server 2 using the communication unit 11B. Send (Sc9). At that time, the message identification unit 154B includes the data indicating the second language stored in the storage unit 14B as the data indicating the language of the translation destination in the voice translation request.

The message identification unit 154B receives the English translation message for voice output and display transmitted from the voice recognition server 2 as a reply to the voice translation request via the communication unit 11B (Sc10). When the translated message is received, the alarm unit 155B outputs the English alarm message by voice from the speaker 17B and displays the English translated message on the display unit 12B (Sc11).

After outputting the translated message, the message identification unit 154B determines whether or not the sensor alert message is output in step Sc11 (Sc12). As a result of this determination, if a sensor alert message is output (YES in Sc12), step Sc1 is executed again. On the other hand, as a result of this determination, if the sensor alert message is not output (NO in Sc12), the message identification unit 154B determines whether or not the fire alarm message is output in step Sc11 (Sc13). .. As a result of this determination, if the fire alarm message is not output (that is, when the non-fire message is output) (NO of Sc13), at least the voice output of the non-fire message is output one or more predetermined times. You may stop after outputting. The operation of stopping the non-fire message may be the same as that of the first embodiment or the second embodiment. That is, to stop the non-fire message, an alarm stop button (not shown) is displayed on the display unit 12B displaying the non-fire message, and the user of the terminal 1B operates the operation unit 13B corresponding to the alarm stop button. By doing so (Sc19), the non-fire message may be stopped (Sc20). At this time, the voice output and display of the non-fire message may be stopped, only the voice output of the non-fire message may be stopped and the display may be left, or the output to be stopped may be appropriately selected. To stop the alarm program 142B, a display unit 12B displaying a non-fire message is displayed with an end button (not shown), and the user of the terminal 1B operates the operation unit 13B corresponding to the end button. (YES of Sc21), the alarm program 142B itself may be stopped (Sc18). On the other hand, if a fire warning message is output as a result of determining whether or not a fire warning message has been output (YES in Sc13), the sweep sound stop recognition unit 157 stops broadcasting the sweep sound for a predetermined time or longer. It is determined whether or not the fire has been performed (Sc14). As a result of this determination, if the sweep sound broadcast is stopped for a predetermined time or longer (in other words, it is determined that the fire broadcast is stopped and the microphone broadcast is performed) (YES of Sc14), at least the fire warning message. The audio output of the above is stopped, and the message identification unit 154B acquires the acoustic signal of the broadcast message to be broadcast after the broadcast of the sweep sound is stopped from the microphone 16B (Sc22). When the acoustic signal is acquired, the message identification unit 154B extracts a feature amount from the acquired acoustic signal (Sc23), and sends a voice translation request including the extracted feature amount data to the voice recognition server 2 using the communication unit 11B. Transmit (Sc24). At that time, the message identification unit 154B includes the data indicating the second language stored in the storage unit 14B as the data indicating the language of the translation destination in the voice translation request.

The message identification unit 154B receives the English translation message for voice output and display transmitted from the voice recognition server 2 as a reply to the voice translation request via the communication unit 11B (Sc25). When the translated message is received, the alarm unit 155B outputs the English translated message by voice from the speaker 17B and displays the English translated message on the display unit 12B (Sc26). After outputting the translated message, step Sc14 is executed again.

As a result of the determination in step Sc14, if broadcasting with a sweep sound (that is, fire broadcasting) continues (NO in Sc14), the fire alarm message is output until the alarm program 142B is stopped, and then repeated. May be good. Further, to stop the fire alarm, an alarm stop button (not shown) is displayed on the display unit 12B during the fire alarm, and the user of the terminal 1B operates the operation unit 13B corresponding to the alarm stop button ( Sc15 YES), the fire alarm may be stopped (Sc16). At this time, the audio output and display of the fire alarm message may be stopped, only the audio output of the fire alarm message may be stopped and the display may be left, or the output to be stopped may be appropriately selected. On the other hand, when the user of the terminal 1B does not operate the operation unit 13B corresponding to the alarm stop button (NO of Sc15), the process proceeds to step Sc17 without stopping the fire alarm. To stop the alarm program 142B, display an end button (not shown) on the display unit 12B during the fire alarm operation, and the user of the terminal 1B operates the operation unit 13B corresponding to the end button (Sc17). YES), the alarm program 142B itself may be stopped (Sc18). If the user of the terminal 1B does not operate the operation unit 13B corresponding to the end button (NO of Sc17), step Sc14 is executed again.
The above is the description of the alarm operation.

According to the terminal 1B described above, as in the terminal 1 according to the first embodiment, even a foreigner who cannot understand Japanese can understand the contents of the emergency broadcast.

3-4. Modification example of the third embodiment The above-mentioned third embodiment may be modified as follows. The following modification may be combined with one or more other modification.

3-4-1. Modification 1
The message identification unit 154B transmits a voice translation request including an acoustic signal of the broadcast message to the voice recognition server 2 instead of the feature amount of the broadcast message broadcast after the signal sound, and features the voice recognition server 2. The amount may be extracted.

3-4-2. Modification 2
The terminal 1B may include the sweep sound recognition unit 153 according to the first embodiment, and may refer to the result of the determination by the sweep sound recognition unit 153 for the identification of the fire broadcast. In that case, the terminal 1B may include the message database 143 according to the first embodiment, and when the fire broadcast is recognized, the message identification unit 154B may acquire the fire warning message from the database.

3-4-3. Modification 3
The terminal 1B may execute all or part of the processing by the voice recognition server 2. For example, all the processing by the voice recognition server 2 may be executed in the terminal 1B, and the processing related to the above-mentioned translation message output operation may be completed in the terminal 1B.

3-4-4. Modification 4
As a reply to the voice translation request, the message identification unit 154B receives a character string representing a broadcast message obtained as a result of voice recognition from the voice recognition server 2, and the character string contains a keyword unique to each broadcast message. The alarm message may be specified by determining whether or not it is included. In that case, the terminal 1B includes a message database that holds each alarm message in association with a keyword unique to the broadcast message corresponding to the alarm message. The message identification unit 154B acquires an alarm message associated with the keyword received from the voice recognition server 2 from this message database.

3-4-5. Modification 5
The terminal 1B assumes that the fire broadcast is stopped and the microphone broadcasts by detecting the stop of the sweep sound broadcast for a predetermined time or longer, and the voice based on the acoustic signal acquired from the microphone 16B. The translation request was sent to the voice recognition server 2, but the present invention is not limited to this. A translation start button (not shown) is displayed on the display unit 12B, and when the operation unit 13B detects an operation corresponding to the translation start button, the above voice translation request is transmitted to the voice recognition server 2 and arbitrarily. You may be able to get the translated message. As a result, after the user of the terminal 1B recognizes the occurrence of a fire, it is possible to arbitrarily obtain a translated message corresponding to the broadcast message. This translation start button may be displayed on the display unit 12B when a fire alarm is output (YES in Sc13). Further, the translation start button may also be used as the alarm stop button. As a result, when the user of the terminal 1B recognizes the fire alarm and performs the alarm stop operation, the translated message corresponding to the broadcast message can be automatically obtained. Further, this translation start button is displayed on the display unit 12B after the alarm program 142B is activated or at other timings so that the user can send the voice translation request to the voice recognition server 2 at an arbitrary timing. Good.

4. Modification example common to the first to third embodiments The above embodiment may be modified as follows. The following modification may be combined with one or more other modification.

4-1. Modification 1
The emergency broadcast assumed in the above embodiment or modification is only an example of disaster broadcast performed by the emergency broadcasting equipment. Disaster broadcasts include any broadcast that guides evacuation from a building. For example, the disaster broadcasting includes "Earthquake Early Warning" linked to the Earthquake Early Warning from the Earthquake Early Warning receiving terminal connected to the emergency broadcasting equipment. The Earthquake Early Warning consists of a signal sound (Pironporone is repeated twice in succession, or Hyun Hyun Hyun is repeated once) and a voice message. The voice message is, for example, a message "It is an earthquake. Please calm down and protect yourself (male voice)" as one unit, and this is repeated twice in succession.

When an earthquake occurs, the microphone according to the above embodiment or modification may pick up the sound of an emergency earthquake broadcast performed in the first language by the emergency broadcasting equipment and output an acoustic signal. In addition, the alarm message identification unit is based on the acoustic pattern indicated by the acoustic signal output by the microphone (specifically, the acoustic pattern of the signal sound or the audio message), and the earthquake early warning message corresponding to the broadcast message of the earthquake early warning broadcast. Therefore, the earthquake early warning message represented by the second language preset by the user of the terminal may be specified. In addition, the alarm program is activated by the activation unit when the earthquake early warning broadcast is detected based on the acoustic pattern indicated by the acoustic signal output by the microphone (specifically, the acoustic pattern of the signal sound). You may do so. In addition, disaster broadcasting from emergency broadcasting equipment is not limited to interlocking with the above-mentioned automatic fire alarm system and Earthquake Early Warning receiving terminal, and various disaster information provided via a communication line (for example, the Internet). (For example, traffic operation information, weather information, disaster / disaster prevention information, J-alert, L-alert, etc.) may be linked. Then, based on the acoustic pattern and characteristic acoustic information peculiar to the signal sound, the voice message, and the like included in the disaster broadcast by these, the message to be notified according to the disaster broadcast may be acquired and notified.

4-2. Modification 2
In the above embodiment or modification, the first language may be English and the second language may be Japanese. In addition to Japanese and English, the combination of the first language and the second language may be arbitrarily selected from natural languages such as German, French, Chinese, and Korean. As another example, when a disaster broadcast is conducted in Japanese, select Japanese for both the first and second languages for the Japanese hearing and visually impaired, and use it in Japanese. The alarm message may be notified to the person. When Japanese is set as the second language of the alarm program, the alarm message is notified in Japanese.

4-3. Modification 3
The terminal according to the above embodiment or modification does not necessarily have to be portable, and may be a stationary computer (for example, digital signage). In this case, the communication unit 11 of the terminal is not limited to wireless communication, and may be connected to a wired LAN. Further, as the second language, one language for audio output and a plurality of languages for simultaneously displaying and outputting in a plurality of languages may be set. Further, as the second language, a plurality of languages may be set for audio output, and audio may be sequentially output in these languages.

4-4. Modification 4
The alarm unit according to the above embodiment or modification may output an alarm message from only one of the speaker and the display unit. For example, the alarm unit can appropriately set the alarm output so that the alarm message is output only to the display unit for the hearing impaired and the alarm message is output only by voice for the visually impaired. It may be set as appropriate according to the degree of disability of the user. Further, for the hearing impaired person, when the alarm message is generated, the vibrator provided in the terminal may be activated to perceive the existence of the alarm message and prompt the user to confirm the display of the alarm message.

4-5. Modification 5
The alarm unit according to the above embodiment or modification may display a pictogram indicating the content of the alarm on the display unit in place of or in addition to the text message as the alarm message for display. For example, in the first and second embodiments, the pictogram image data is associated with the display text message, or the pictogram image data is associated with the display text message, and the message database 143 is used as a display warning message. , 143A may be held. Alternatively, instead of these pictogram image data, the pictogram identification information is retained, and the corresponding pictogram image data is acquired by referring to a database (not shown) in which the identification information and the pictogram image data are stored in association with each other. You may do so. In the third embodiment, it is preferable to give the pictogram to be displayed by the above identification information based on the unique keyword included in the translated message.

4-6. Modification 6
The alarm program according to the above embodiment or modification may be provided via a recording medium that can be read by a terminal using computing technology. Here, the recording medium is, for example, a magnetic recording medium such as a magnetic tape or a magnetic disk, an optical recording medium such as an optical disk, an optical magnetic recording medium, a semiconductor memory, or the like. Further, this alarm program may be provided via a network such as the Internet. Further, the alarm program may include the above-mentioned automatic start program.

1, 1A, 1B ... Terminal, 2 ... Voice recognition server, 11, 11A, 11B ... Communication unit, 12, 12A, 12B ... Display unit, 13, 13A, 13B ... Operation unit, 14, 14A, 14B ... Storage unit, 15, 15A, 15B ... Control unit, 16, 16A, 16B ... Microphone, 17, 17A, 17B ... Speaker, 21 ... Communication unit, 22 ... Storage unit, 23 ... Control unit, 141, 141A, 141B ... Automatic start program, 142, 142A, 142B ... alarm program, 143, 143A ... message database, 151, 151A, 151B ... signal sound recognition unit, 152, 152A, 152B ... activation unit, 153 ... sweep sound recognition unit, 154, 154A ... alarm message identification Unit, 154B ... Message identification unit, 155, 155A, 155B ... Alarm unit, 156 ... Broadcast message identification unit, 157 ... Sweep sound stop recognition unit, 231 ... Voice translation request reception unit, 232 ... Voice recognition unit, 233 ... Machine translation Unit, 234 ... Voice synthesis unit, 235 ... Message transmission unit

Claims (3)

A disaster broadcast that is performed in the first language by emergency broadcasting equipment when a disaster occurs, and has an acoustic input unit that collects the sound of the disaster broadcast consisting of a warning sound and a voice message and outputs an acoustic signal, and an alarm message. A terminal using computing technology equipped with an alarm output unit that outputs
An alarm message corresponding to the broadcast message of the disaster broadcast based on the acoustic pattern of the warning sound indicated by the acoustic signal output by the acoustic input unit, which is preset by the user of the terminal. An alarm message identification unit that identifies an alarm message expressed in two languages,
An alarm program for functioning as an alarm unit that outputs an alarm message specified by the alarm message identification unit to the alarm output unit. The alarm program according to claim 1, which is activated when the disaster broadcast is detected based on the acoustic pattern of the warning sound indicated by the acoustic signal output by the acoustic input unit. A disaster broadcast that is performed in the first language by emergency broadcasting equipment in the event of a disaster, and has an acoustic input unit that collects the sound of the disaster broadcast consisting of warning sounds and voice messages and outputs an acoustic signal.
An alarm output unit that outputs an alarm message and
An alarm message corresponding to the broadcast message of the disaster broadcast based on the acoustic pattern of the warning sound indicated by the acoustic signal output by the acoustic input unit, which is preset by the user of the own terminal. An alarm message identification unit that identifies an alarm message expressed in two languages,
A terminal including an alarm unit that outputs an alarm message specified by the alarm message identification unit to the alarm output unit.

Images