JP7288619B1 - elevator system - Google Patents

Abstract

Kind Code: A1 A predetermined voice in a language that the user is likely to recognize can be notified to the user first, thereby shortening the time the user feels uneasy and contributing to reassurance of the user. can. Kind Code: A1 An elevator system for informing a user in a car moving on a moving route of a predetermined guidance voice, comprising voice storage means for storing the predetermined guidance voice in a plurality of languages, A notification order is set in advance for each language, and feature collecting means for collecting information including features that appear to the user in the car; and a language used by the user among the features. feature extracting means for extracting relevant identification features; notification order renewing means for renewing the notification order; and notification of the predetermined guidance voice in the renewed notification order using the plurality of languages. and notification means, wherein the notification order renewing means renews the notification order in accordance with the identification feature. [Selection drawing] Fig. 1

Description

The present invention relates to elevator systems.

For example, Patent Literature 1 exists as a document that discloses a device for providing guidance in an elevator car by means of voice broadcasting. In this patent document 1, assuming that the passengers of the elevator are foreigners, it is described that guidance is provided in a foreign language as a guidance broadcast from a speaker.

By the way, there are a wide variety of foreigners who use elevators, and each foreigner speaks a variety of languages. For this reason, conventional elevators have adopted a method of sequentially announcing pre-recorded broadcast content in each language.

JP-A-6-1549

However, in the above-described conventional method, since the broadcast contents in each language are announced in a fixed order, the user has to wait until a language that they can understand is announced. It made people feel uncomfortable. In addition, for example, in the event of an emergency such as an accident or disaster, in situations where users are trapped in an elevator, if content related to their safety is not announced in a language they can understand, it can make users feel uneasy. could be connected.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to contribute to reassurance of users using elevators with respect to guidance by voice broadcasting in an emergency.

The present invention is an elevator system for informing a user in a car moving on a moving route of a predetermined guidance voice, comprising voice storage means for storing the predetermined guidance voice in a plurality of languages, A notification order is set in advance for each language, and feature collecting means for collecting information including features that appear to the user in the car; and a language used by the user among the features. feature extracting means for extracting relevant identification features; notification order renewing means for renewing the notification order; and notification of the predetermined guidance voice in the renewed notification order using the plurality of languages. and notification means, wherein the notification order renewal means renews the notification order in accordance with the identification feature.

According to the above configuration, the notification order renewal means renews the notification order in accordance with the identification feature extracted by the feature extraction means from among the information collected by the feature collection means. can notify the user of the predetermined guidance voice stored in the voice storage means, with the language likely to be recognized by the user being given first.

Further, in the present invention, the identification features include the user's appearance features and the user's linguistic features at the time of utterance, and the notification order renewal means is configured to update the notification order related to the renewal. A plurality of order determination means for determining languages one by one, wherein one or more of the order determination means among the plurality of order determination means determines the language according to the appearance characteristics, One or more of said order determining means different from each order determining means may determine said language according to said linguistic feature.

According to the above configuration, when the notification order renewing means renews the notification order, one or more of the order determination means determines the language according to the appearance characteristics, and the order determination means Since one or more of the order determining means different from each determine the language according to the linguistic feature, the language of each notification order can be determined from various viewpoints.

Further, in the present invention, the appearance feature may be a feature related to the user's body or clothing, and the feature collecting means may include an image capturing means for capturing an image of at least the user.

According to the above configuration, as part of collecting the information, the external features can be collected by capturing an image of the user with the imaging means.

Further, in the present invention, the feature collection means collects speech based on the user's utterance, and the feature extraction means extracts a word or a plurality of words having a certain meaning from the speech as the linguistic features. A set of words may be extracted.

According to the above configuration, a word or a set of words having a certain meaning can be extracted as the linguistic feature from the voice uttered by the user. can be used to extract the identification feature, and the notification order updating means can update the preset notification order.

Further, in the present invention, the feature extraction means may convert voice data based on the user's utterance into character data, and extract the linguistic features from the character data.

According to the above configuration, since the linguistic features are extracted from the converted character data, for example, words and sentences extracted from the character data can be used in renewing the notification order by the notification order renewing means. .

Further, in the present invention, the feature collecting means may collect the voice of the user's inquiry using an external communication device in the car.

According to the above configuration, it is possible to renew the notification order by using the voice related to the inquiry from the user.

Therefore, in the present invention, since the user can be notified of the predetermined voice in a language likely to be recognized by the user first, the time during which the user feels uneasy can be shortened, and the user can be relieved. can contribute to

FIG. 1 is a system configuration diagram of one embodiment of the elevator system of the present invention. FIG. 2 is a flow chart showing the overall processing of the elevator system. FIG. 3 is a flow chart showing processing by the first means in the elevator system. FIG. 4 is a flow chart showing processing in the second means in the elevator system. FIG. 5 is a flow chart showing the process when the answer to step 27 in FIG. 4 is Yes. FIG. 6 is a flow chart showing processing by the third means in the elevator system. FIG. 7 is a flow chart showing the process when the answer to step 37 in FIG. 6 is Yes. FIG. 8 is a flow chart showing processing in the fourth means in the elevator system. FIG. 9 is a flow chart showing the process when the answer to step 47 in FIG. 8 is Yes.

An elevator system according to the present invention will be described below. Elevator systems, for example, in situations where users are confined in a car due to unforeseen circumstances such as disasters and accidents, can quickly provide guidance in a language that users can understand, thereby reducing their anxiety and impatience. used to relieve and reassure This elevator system is for grasping the user's recognizable language in order to promptly transmit information to the user.

This elevator system includes voice storage means for storing predetermined guidance voices to be conveyed to users in a plurality of languages, notification order updating means for renewing the order of notifying languages, and information including features that appear to users. feature collecting means for collecting, feature extracting means for extracting identification features related to the language used by the user from among the features, and notification means for notifying the user of a predetermined guidance voice .

In the elevator system 1 of the present embodiment, the user's recognizable language is grasped by exchanging information between the inside of the car 2 and the external device 3 . Therefore, the elevator system 1 of this embodiment includes a car 2 and an external device 3 provided separately from the car 2, as shown in FIG. In this embodiment, the car 2 and the external device 3 exchange information via the communication system R. The communication system R is, for example, a communication system capable of two-way communication between the processing microcomputer 26 and the server 32 via a public line (telephone line or Internet line). It should be noted that the external device 3 of the present embodiment is assumed to be provided in a place (for example, a monitoring center) different from the building in which the car 2 is provided.

The car 2 moves in a hoistway as a movement path with people and goods on it. The car 2 includes an operation panel 21 having various operation buttons 24 for operating the car 2, such as a destination registration button for registering a desired floor and a door opening/closing button for opening/closing a car door (not shown), A monitoring camera 20 for monitoring the inside of the car 2 is provided. The operation panel 21 is provided with a display unit 23 for displaying operation information of the car 2, such as the number of floors the car 2 is passing through or stopped, and the moving direction of the car 2.

By the way, if an unexpected situation occurs in the elevator, the users in the car 2 need to inform the outside of the situation inside the car 2 or ask for help. Therefore, the car 2 is provided with an external communication device 22 (specifically, an intercom) for communicating with the outside. The external communication device 22 has an emergency call button 220 for activating the external communication device 22, a microphone 221 and a speaker 222 functioning as communication means. For example, pressing the emergency call button 220 activates the external communication device 22 . Specifically, when the external communication device 22 is activated, the microphone 221 can collect the voice of the user in the car 2 and the speaker 222 can output the voice. Then, communication with the external device 3 is started using the processing microcomputer 26, which will be described later. Therefore, it is possible to talk with an outside person (for example, an operator stationed at a monitoring center that monitors elevators). Incidentally, in this embodiment, the external communication device 22 is provided on the operation panel 21 .

The car 2 is provided with notification means for notifying the users in the car 2 . In this embodiment, the speaker 222 provided in the external communication device 22 corresponds to the notification means. Therefore, the notification means of this embodiment is provided on the operation panel 21 .

Inside the car 2, there is provided a feature collecting means for collecting information on the user. This feature collection means collects information including features that appear to the users in the car 2 . Here, the "features that appear in the user" indicate the individual user. For example, linguistic features such as the voice when the user speaks; Appearance features of the user such as clothing and accessories (clothes) correspond.

The feature collection means of the present embodiment includes an imaging means for imaging the inside of the car 2 in order to collect appearance features of the user such as face, physique such as height and body type, and clothing. For example, the monitoring camera 20 provided on the ceiling inside the car 2 corresponds to the imaging means. Further, the operation panel 21 of this embodiment is provided with a camera 25 as imaging means. This imaging means images the user. Further, the feature collection means of this embodiment includes sound collection means for collecting sounds in the car 2 in order to collect the linguistic features of the user. This sound collecting means collects the voice of the user in the car 2 (for example, the voice of the user speaking into the microphone 221 or the conversation between the users in the car 2). In this embodiment, the microphone 221 provided in the external communication device 22 corresponds to the sound collecting means. Therefore, the feature collecting means of the present embodiment captures an image of the user with the imaging means and collects the voice of the user with the sound collecting means, thereby collecting information on the features that appear to the user from various angles. there is Further, in this embodiment, since the imaging means and the sound collecting means are provided inside the car 2 , the feature collecting means directly collects information about the users in the car 2 .

A processing microcomputer 26 is incorporated in the operation panel 21 of this embodiment. The processing microcomputer 26 issues a command to open and close the door when the door open/close button is pressed, and issues a command to the car 2 to move to a predetermined floor when the destination registration button or the call button provided at each platform is pressed. It is for When the emergency call button 220 is pressed, the processing microcomputer 26 also issues a command to the external communication device 22 to activate it. Further, the processing microcomputer 26 issues a command to the notification means to notify the guidance voice received from the external device 3, which will be described later.

When the emergency call button 220 is pressed, the processing microcomputer 26 of this embodiment issues a command to the feature collecting means to collect information. As a result, imaging by the imaging means and sound collection by the sound collecting means are started. Therefore, in this embodiment, when the call button is pressed, the processing microcomputer 26 is configured to issue a command to the external communication device 22 and a command to the feature collection means.

The processing microcomputer 26 is connected via a communication system R to be communicable with a server 32, which will be described later. Therefore, the user can communicate with the outside by using the external communication device 22 and the processing microcomputer 26 . Therefore, in this embodiment, the processing microcomputer 26 sends the information collected by the feature collection means (specifically, the sound collected by the sound collection means and the image inside the car 2 captured by the imaging means) to the server 32. configured to send to

The external device 3 is connected to the car 2 via a communication system R. The external device 3 has a storage device 30 , a database 31 and a server 32 . The external device 3 is arranged, for example, in a monitoring center that constantly monitors the operating state of the elevator. An operator is permanently stationed at this monitoring center, and responds to communication by the external communication device 22 from the user in the car 2. - 特許庁

The storage device 30 stores a predetermined guidance voice. Here, the "predetermined guidance voice" is for notifying the user in the car 2 of the information. This "predetermined guidance voice" is a language that the user can recognize, such as "Can you understand this word? Raise your hand if you can." In order to understand, the user calls out to the user after grasping the user's recognizable language such as "please wait for a while" or "a rescuer will arrive in about 5 minutes." It corresponds to information transmission voice. The storage device 30 of this embodiment stores various types of guidance voices. Further, the storage device 30 stores guidance voices in a plurality of languages. Specifically, the storage device 30 of this embodiment stores guidance voices in four languages: Japanese, English, Korean, and Chinese. This storage device 30 functions as voice storage means.

The database 31 has basic information that serves as a basis for grasping the user's comprehensible language. Here, "basic information" corresponds to, for example, linguistic information such as words and sentences in various languages such as English, Japanese, Chinese, and Korean. In addition, in the "basic information", in addition to characteristics that are said to appear in a high percentage of people in each country, such as eye color, face length, and other facial characteristics, clothing characteristics that correspond to ethnic groups , For example, covering from head to neck with soft things (Indonesia's "hijab" etc.), for example, whether the clothes are flashy or plain, lightly dressed, or wearing school uniforms It includes customary appearance information such as Basic information is associated with each language. Appearance features are associated with each likely corresponding language.

The server 32 is connected to the processing microcomputer 26 via the communication system R. The server 32 uses the information received from the processing microcomputer 26 to determine the order in which the languages are notified, and issues a command to the processing microcomputer 26 to perform notification in the determined order. This server 32 is connected to the database 31 and the storage device 30 . The server 32 presets the notification order for the languages set in the storage device 30 . Therefore, an initial notification order is set in advance for the languages stored in the storage device 30 .

The feature extraction means extracts identification features from information such as images and voices including features that appear to the user sent from the processing microcomputer 26 . Specifically, the identification feature is a feature for grasping a language that the user can recognize, and corresponds to, for example, a linguistic feature or an appearance feature.

The notification order updating means updates a preset (temporary) notification order (initial notification order) to another notification order (renewed notification order). The notification order renewal means renews the notification order according to the identification feature. In this embodiment, the notification order is renewed by determining the language in ascending order using the distinguishing features. Therefore, the notification order renewing means includes a plurality (four) of each order determining means for determining the language of the renewing notification order one by one.

Each order determining means determines the language from various points of view. In this embodiment, there is a high possibility that the first order determining means will determine what the user can recognize as the language, and as the second, third, and fourth order progresses, what the user can recognize as the language Designed to make decisions less likely. Therefore, the languages determined by the second to fourth order determining means have meanings determined auxiliary to the language determined by the first order determining means. Further, in the present embodiment, each of the first to fourth order determination means corresponds to each order in the notification order.

The first order determination means of this embodiment determines the language according to the linguistic features. Specifically, the first order determination means determines the language using the sound information sent from the processing microcomputer 26 and the basic information stored in the database 31 . Second to fourth order determining means determine the language according to the appearance features. Specifically, the second to fourth order determination means determine the language using the image information sent from the processing microcomputer 26 and the basic information stored in the database 31 . On the other hand, the second-order determining means allocates ethnic characteristics such as the user's clothing and accessories as language determination. The third-order determining means roughly estimates the ethnic group from the user's facial features to determine the language, and assigns a language that has a high percentage of use in that ethnic group. The fourth-ranking means determines the language by roughly estimating the region based on the user's habitual characteristics such as the user's clothes, and assigning the languages with the highest percentage of use in the region.

Next, with reference to FIGS. 2 to 9, the flow of grasping the language that the user can recognize in the elevator system 1 according to this embodiment will be described.

First, assume that the car 2 is stopped due to an unforeseen event such as a disaster, and the user is confined in the car 2 . At this time, many users are expected to use the external contact device 22 provided on the control panel 21 to try to make an inquiry to the outside. Specifically, the user in this case activates the external communication device 22 by pressing the emergency call button 220, and speaks into the microphone 221 in a language that the user can use.

On the other hand, when the emergency call button 220 is pressed, the elevator system 1 of the present embodiment starts processing for grasping the language that the user can recognize. command to collect information that includes features that appear to the user. This causes the feature collecting means to start collecting information. Specifically, information is collected by capturing images of the inside of the car 2 using the camera 25 and the monitoring camera 20 as imaging means, and collecting sounds inside the car 2 using the microphone 221 . Therefore, in the present embodiment, the monitor camera 20 and the camera 25 capture an image of the user, and the microphone 221 collects the voice uttered by the user. Then, the processing microcomputer 26 sends the information collected by the feature collection means to the server 32 via the communication system R.

In the server 32 that has received the information, the feature extraction means extracts the identification feature from the information, and the notification order renewal means renews the notification order according to the identification feature. As shown in FIG. 2, in this embodiment, each of the first to fourth order determination means determines the language (steps S1, S2, S3, S4), and the notification order renewal means renews the initial notification order. Renewed to notification order. First, the process of determining the first language in the new notification order (step S1) will be described.

The server 32 sets the first language in the initial setting order as the first determined language (step S10). Server 32 then performs speech recognition on the sound information using the first determined language. As speech recognition, first, the user's speech is extracted from sound information. Then, words or a set of words having a certain meaning are extracted from the speech. Specifically, the sound data is converted into character data, and words and sentences containing words are extracted from the character data as linguistic features (step S11). Then, the word or the sentence containing the word is compared with the basic information related to the language set as the first decision language among the basic information recorded in the database 31 (step S12), and it is confirmed whether they match. (Step S13). Then, if they match (step S13: Yes), the first determined language is determined to be the first language in the renewal notification order (step S14).

On the other hand, if it is compared with the basic information and does not match (step S13: No), the server 32 sets the second language, which is next in the initial setting order, as the first decision language (step S15). ). Thereafter, it is checked whether speech recognition has been performed in all languages, and if not (step S16: No), the linguistic features are compared with the basic information again to search for a matching language (step S13). . In this way, the server 32 sequentially sets the first determined language from the first in the initial setting order and performs speech recognition, thereby determining the first language in the update notification order. On the other hand, if speech recognition is performed in all languages and the linguistic features do not match the basic information (step S16: Yes), the first language in the initial setting order is changed to the first in the renewal notification order. language is determined (step S17).

Next, the process of determining the second language (step S2) will be described. When determining the second language, first, the first language in the initial setting order is set as the second determined language (step S20). After that, it is confirmed whether the language set as the second determined language has been determined as the first language in the update notification order (step S21). If the second determined language is not determined to be the first language in the renewal notification order (step S21: No), image recognition is performed using the second determined language. For image recognition, among the information received from the processing microcomputer 26, from the image information, for example, the outline is straight and hard, the outline width is more than a certain value, so it is thick, and the composition line (wrinkle) flows horizontally. A variety of appearance features such as a wound or a vertical running wrinkle is extracted, and the language is determined by the combination of the appearance features. In addition, the appearance characteristics referred to in the second language determination process indicate characteristics related to clothing such as ethnic clothing, which are characteristic depending on the area of residence. That is, the feature extraction means extracts, from the image information, the appearance features indicating the national costume associated with the language set as the second decision language (step S22), and the appearance features extracted by the feature extraction means are Among the information, the language set as the second decision language is compared with information such as folk costumes related (step S23) to confirm whether they match (step S24). Then, if they match (step S24: Yes), the second determined language is determined to be the first language in the renewal notification order (step S25).

On the other hand, if the second determined language is determined to be the first language in the renewal notification order (step S21: Yes), or if the user is wearing a suit or necktie, for example, the appearance without individuality However, if the appearance features do not match the basic information (step S24: No), the second language following the initial setting order is set as the second determined language (step S26). After that, it is confirmed whether image recognition has been performed in all languages, and if not (step S27: No), the language set as the second decision language is again decided as the first language in the renewal notification order. (Step S21), the feature extraction means extracts appearance features related to the language set as the second decision language from the image information, and attempts to compare them with the basic information (Steps S22, S23). In this way, the server 32 sequentially sets the second determined language from the first in the initial setting order and performs image recognition, thereby determining the second language in the renewal notification order.

By the way, when image recognition is performed in all languages and the appearance features do not match the basic information (step S27: Yes), the first language in the initial setting order is the first language in the renewal notification order. It is determined whether the language is determined (step S28), and the first language or the second language in the initial setting order is determined as the second language in the update notification order (steps S280, S281).

Next, the process of determining the third language (step S3) will be described. In order to determine the third language, first, the first language in the initial setting order is set as the third determined language (step S30). Then, it is confirmed whether the language set as the third determined language has been determined as the first or second language in the renewal notification order (step S31). Then, if the first or second language has not been determined (step S31: No), image recognition is performed using the third determined language. Concretely, the feature extracting means, from the image information sent from the processing microcomputer 26, the appearance indicating the feature of the user's face that is considered to be highly likely to be related to the language set as the third decision language characteristic features are extracted (step S32). Extraction of appearance features is carried out by checking, for example, whether the color of the eyes is good and whether the cheeks can be clearly distinguished. Then, the features of the user's face extracted as appearance features are compared with facial information considered to be highly likely to be associated with the language set as the third decision language among the basic information (step S33). , if they match (step S34: Yes), the ethnic group is roughly estimated, and the third language in the new notification order is determined as the language with the highest usage rate in that ethnic group (step S35). ).

On the other hand, if the language set as the third determined language is the language determined as the first or second language in the renewal notification order (step S31: Yes), or if the appearance features and the basic information are the same If they do not match (step S34: No), the second language following the initial setting order is set as the third determined language (step S36). After that, it is checked whether the image recognition has been performed in all the languages, and if not (step S37: No), the language set as the third decision language is again the first or second language in the renewal notification order. is determined (step S31), and the feature extracting means determines from the image information the appearance indicating the features of the user's face that are likely to be associated with the language set as the third determined language. characteristic features are extracted (step S32). In this way, the server 32 sequentially sets the third determined language from the first in the initial setting order and performs image recognition, thereby determining the third language.

By the way, image recognition is performed in all languages, and if the user's facial features, which are appearance features, cannot be extracted from the image information (step S37: Yes), the first or second initial setting order It is determined whether the language is determined to be the first or second language in the new notification order (steps S38, S39), and one of the first to third languages in the initial setting order is selected as the new notification order. is determined as the third language (steps S380, S390, S391).

Next, the fourth language determination process (step S4) will be described. In order to determine the fourth language, first, the first language in the initial setting order is set as the fourth determined language (step S40). Then, it is confirmed whether the language set as the fourth determined language has been determined as one of the first to third languages in the renewal notification order (step S41). Then, if one of the first to third languages has not been determined (step S41: No), image recognition is performed using the fourth determined language. Specifically, the feature extracting means extracts, from the image information sent from the processing microcomputer 26, appearance features indicating user habits associated with the language set as the fourth determined language (step S42). . Here, the extraction of the appearance features that indicate the habits of the user includes, for example, whether the user's clothes are generally flashy or plain, whether the user's clothes are lightly dressed, This is done by confirming items such as whether the person is wearing a school uniform. Then, the features of the extracted appearance features indicating the habits of the user are compared with the information about habits that is considered to be highly likely to be related to the language set as the fourth decision language among the basic information (step S43). ), if they match (step S43: Yes), the region is roughly estimated, and the fourth language is determined to be the fourth language in the renewal notification order as the language with the highest usage rate in that region (step S45).

On the other hand, when the language set as the fourth determined language is the language determined as one of the first to third languages (step S41: Yes), the appearance features and the basic information do not match. If not (step S44: No), the second language next to the initial setting order is set as the fourth determined language (step S46). After that, it is checked whether image recognition has been performed in all languages, and if not applicable (step S47: No), again, the language set as the fourth decision language is one of the first to third in the renewal notification order. It is confirmed whether one language is determined (step S41), and the feature extracting means extracts from the image information the habits of the user that are likely to be associated with the language set as the fourth determined language. The external features shown are extracted (step S42). In this way, the server 32 sequentially sets the fourth determined language from the first in the initial setting order and performs image recognition, thereby determining the fourth language.

Image recognition is performed in all languages, and if the appearance features indicating the user's habits cannot be extracted from the image information (step S47: Yes), the first to third languages in the initial setting order are renewed. It is determined whether one of the first to third languages in the notification order has been selected (steps S48, S49, S50), and one of the first to fourth languages in the initial setting order is changed to the new notification order. is determined as the fourth language (steps S480, S490, S500, S501).

Each of the first to fourth order determination means determines the language (steps S1, S2, S3, S4), and when the notification order renewal means renews the initial notification order to the renewed notification order, the communication system R is changed. A command is sent from the server 32 to the processing microcomputer 26 via the server 32 to perform notification in the new notification order with the guidance voice. out. Then, the notification means notifies the guidance voices in the renewal notification order.

When announcing in the new notification order, for example, the notification means uses the second language in the new notification order, and an interrogation voice (e.g., “Do you understand this word? or "Please stay away from the door"), and when the user takes action in accordance with the guidance voice, the processing microcomputer 26 selects the second language in the new notification order as the user's recognizable language. After that, only the second language in the new notification order is used for notification by the notification means. Further, in the notification by the notification means after recognizing the language as the user's recognizable language, information transmission voice is notified. Therefore, if the language can be recognized as a language that the user can recognize, even if it is in the middle of the renewal notification order, subsequent notification in another language and voice guidance using the calling voice will not be performed.

As described above, according to the present embodiment, the feature extraction means extracts the linguistic features and appearance features as identification features from among the information collected by the feature collection means, and the first to fourth order determination means are selected. Since each of the order determination means determines the language, the notification order renewing means renews the notification order. A user can be notified of the stored predetermined guidance voice. Specifically, in this embodiment, the language order is renewed (replaced) so that the language order is changed from the one with the highest possibility of being recognized by the user to the one with the lowest possibility. A predetermined guidance voice can be notified to the person.

Further, in this embodiment, when the notification order renewal means is renewed to the renewal notification order, the first order determination means determines the language according to the linguistic characteristics, and the second to fourth order determination means , since the language is determined according to the appearance features, the language for each notification order can be determined from various viewpoints. In particular, in each of the second to fourth order determining means, the feature extracting means extracts appearance features that indicate clothes, the user's face, and habits, so that the language can be determined by finely classifying the appearance features.

Further, in this embodiment, it is highly likely that the first order determination means will determine the language recognizable by the user. Since it is configured so that the possibility of determining a thing is low, the renewal notification order can be set in the order of high possibility of recognition by the user.

It should be noted that the present invention is not limited to the above embodiments, and can be modified as appropriate without departing from the gist of the invention.

In the above-described embodiment, the speaker 222 only outputs a guidance voice message, but in addition to the above, for example, the display unit 23 may display text messages in order of renewal notification. .

In the above embodiment, the external communication device 22 is provided with the emergency call button 220. However, in order to make the emergency call button 220 recognizable by people in languages other than Japanese and English, the emergency call button 220 is provided with, for example, A pictogram display may be made.

In the above embodiment, the automatic response system for outputting the standard message from the server 32 side may be configured to output the standard message from the processing microcomputer 26, for example. Also, when a standard message is to be sent from the server 32, the operator can set the standard message for the server 32. FIG.

In the above-described embodiment, a case has been described in which the elevator system 1 is used in order to grasp the language that the user can understand before performing voice guidance. At this stage, the elevator system 1 may be used to grasp the language used by the user.

In the above-described embodiment, as speech recognition in the process of determining the first language in the order of renewal notification, words, sentences containing words, and basic information recorded in the database 31 are set as the first decision language. The basic information related to the language was compared, but not limited to this, the voice data is converted into text in multiple languages, the text data and the basic information are compared, and the most matching one is selected as the first language. can be set to

In the above embodiment, the external device 3 is installed in a place different from the building where the car 2 is installed. control room above the hoistway as a travel path).

In the elevator system 1 of the above-described embodiment, a case has been described in which the inside of the car 2 and the external device 3 grasp the language that the user can recognize by exchanging information via the communication system R. and the external device 3 (that is, no sound or image is sent from the processing microcomputer 26 to the server 32). good. In this case, for example, the processing microcomputer 26 may be provided with voice storage means, notification order renewal means, and feature extraction means. Specifically, the processing microcomputer 26 has basic information for comparison with predetermined guidance voices, appearance features, and linguistic features, and the processing microcomputer determines the language order. Further, in this case, in addition to the automatic response from the processing microcomputer 26, the automatic response at the monitoring center or the response by the operator may be used.

In the above embodiment, the server 32 automatically sends the selected guidance voice to the processing microcomputer 26. For example, the guidance voice sent from the server 32 to the processing microcomputer 26 may You may choose

Further, in the above-described embodiment, when the emergency call button 220 is pressed, the elevator system 1 of the present embodiment starts the process of grasping the language that the user can recognize. Specifically, the processing microcomputer 26 A case has been described where a command is issued to the feature collection means to collect information including features that have appeared to the user. 2, the elevator system 1 may start the process of grasping the user's recognizable language on condition that the processing microcomputer 26 detects that the elevator system 2 has come to an emergency stop. In this case, when a plurality of users are talking in the car 2, for example, the voice of the person closest to the microphone 221 is extracted from the sound information, or the voice with the highest volume among the voices collected by the microphone 221 is extracted. A loud sound may be extracted.

1: elevator system, 2: car, 20: surveillance camera, 21: control panel, 22: external communication device, 220: emergency call button, 221: microphone, 222: speaker, 23: display unit, 24: operation button, 25: camera, 26: processing microcomputer, 3: external device, 30: storage device, 31: database, 32: server, R: communication system

Claims (6)

An elevator system for notifying a user in a car moving on a moving route with a predetermined guidance voice,
Voice storage means for storing the predetermined guidance voice in a plurality of languages,
The order of notification is set in advance for each of the languages,
a feature collecting means for collecting information including features that appear to the user in the car;
Features for extracting distinguishing features related to the language used by the user from among the features, including features of the user's appearance and linguistic features when the user speaks. extraction means;
a notification order renewing means for renewing the notification order;
a notification means for notifying the predetermined guidance voice using the plurality of languages in the renewed notification order;
The elevator system, wherein the notification order renewal means renews the notification order in accordance with the identification feature. An elevator system for notifying a user in a car moving on a moving route with a predetermined guidance voice,
Voice storage means for storing the predetermined guidance voice in a plurality of languages,
The order of notification is set in advance for each of the languages,
a feature collecting means for collecting information including features that appear to the user in the car;
Feature extracting means for extracting an identification feature related to the language used by the user from among the features;
a notification order renewing means for renewing the notification order;
a notification means for notifying the predetermined guidance voice using the plurality of languages in the renewed notification order;
The notification order renewal means renews the notification order according to the identification feature,
The identification features include appearance features of the user and linguistic features of the user's utterance,
The notification order renewal means comprises a plurality of order determination means for determining the language of the notification order related to the renewal one by one,
One or more of the order determining means out of the plurality of order determining means determines the language according to the appearance features, and one or more of the order determining means different from the order determining means. An elevator system, wherein means determines said language in response to said linguistic features. The appearance features are features related to the user's body or clothing,
3. The elevator system of claim 2, wherein said feature collecting means comprises imaging means for imaging at least said user. The feature collection means collects voice based on the user's utterance,
3. The elevator system according to claim 2, wherein said feature extracting means extracts a word or a set of words having a certain meaning from said speech as said linguistic features. 5. The elevator system according to claim 4, wherein said feature extraction means converts voice data based on said user's utterance into character data, and extracts said linguistic features from said character data. 6. The elevator system according to claim 4 or 5, wherein said feature collection means collects the voice of said user's inquiry using an external communication device in said car.

Images